Multiplex imaging is revolutionizing pathology by enabling the simultaneous visualization of multiple biomarkers within tissue samples, providing molecular-level insights that traditional hematoxylin and eosin (H&E) staining cannot provide. However, the complexity and cost of multiplex data acquisition have hindered its widespread adoption. Additionally, most existing large repositories of H&E images lack corresponding multiplex images, limiting opportunities for multi-modal analysis. To address these challenges, we leverage recent advances in latent diffusion models (LDMs), which excel at modeling complex data distributions by utilizing their powerful priors for fine-tuning to a target domain. In this paper, we introduce a novel framework for virtual multiplex staining that utilizes pretrained LDM parameters to generate multiplex images from H&E images using a conditional diffusion model. Our approach enables marker-by-marker generation by conditioning the diffusion model on each marker, while sharing the same architecture across all markers. To tackle the challenge of varying pixel value distributions across different marker stains and to improve inference speed, we fine-tune the model for single-step sampling, enhancing both color contrast fidelity and inference efficiency through pixel-level loss functions. We validate our framework on two publicly available datasets, notably demonstrating its effectiveness in generating up to 18 different marker types with improved accuracy, a substantial increase over the 2-3 marker types achieved in previous approaches. This validation highlights the potential of our framework, pioneering virtual multiplex staining. Finally, this paper bridges the gap between H&E and multiplex imaging, potentially enabling retrospective studies and large-scale analyses of existing H&E image repositories.

We present ViSTAR, a Virtual Skill Training system in AR that supports self-guided basketball skill practice, with feedback on balance, posture, and timing. From a formative study with basketball players and coaches, the system addresses three challenges: understanding skills, identifying errors, and correcting mistakes. ViSTAR follows the Behavioral Skills Training (BST) framework-instruction, modeling, rehearsal, and feedback. It provides feedback through visual overlays, rhythm and timing cues, and an AI-powered coaching agent using 3D motion reconstruction. We generate verbal feedback by analyzing spatio-temporal joint data and mapping features to natural-language coaching cues via a Large Language Model (LLM). A key novelty is this feedback generation: motion features become concise coaching insights. In two studies (N=16), participants generally preferred our AI-generated feedback to coach feedback and reported that ViSTAR helped them notice posture and balance issues and refine movements beyond self-observation.

Training resources for parasports are limited, reducing opportunities for athletes and coaches to engage with sport-specific movements and tactical coordination. To address this gap, we developed BRIDGE, a system that integrates a reconstruction pipeline, which detects and tracks players from broadcast video to generate 3D play sequences, with an embodiment-aware visualization framework that decomposes head, trunk, and wheelchair base orientations to represent attention, intent, and mobility. We evaluated BRIDGE in two controlled studies with 20 participants (10 national wheelchair basketball team players and 10 amateur players). The results showed that BRIDGE significantly enhanced the perceived naturalness of player postures and made tactical intentions easier to understand. In addition, it supported functional classification by realistically conveying players’ capabilities, which in turn improved participants’ sense of self-efficacy. This work advances inclusive sports learning and accessible coaching practices, contributing to more equitable access to tactical resources in parasports.

Large language models demonstrate impressive performance on downstream tasks, yet requiring extensive resource consumption when fully fine-tuning all parameters. To mitigate this, Parameter Efficient Fine-Tuning (PEFT) strategies, such as LoRA, have been developed. In this paper, we delve into the concept of task-specific directions (TSDs)-critical for transitioning large models from pretrained states to task-specific enhancements in PEFT. We propose a framework to clearly define these directions and explore their properties, and practical utilization challenges. We then introduce a novel approach, LoRA-Dash, which aims to maximize the impact of TSDs during the fine-tuning process, thereby enhancing model performance on targeted tasks. Additionally, based on our exploration of TSD, we focus on an important issue in PEFT: the initialization of LoRA. While some works have pointed out the significance of initialization for LoRA's performance and proposed various strategies, these methods are often empirical and not task-specific. To address this issue, we propose LoRA-Init. Starting from TSD, we identify the directions that require the most adjustment during fine-tuning for downstream tasks. By initializing the matrices in LoRA with these directions, LoRA-Init significantly enhances LoRA's performance. Moreover, we can combine LoRA-Dash and LoRA-Init to create the final version of LoRA based on TSDs, which we refer to as LoRA-TSD. Extensive experiments have conclusively demonstrated the effectiveness of these methods, and in-depth analyses further reveal the underlying mechanisms of these methods.

We present a data-domain sampling regime for quantifying CNNs' graphic perception behaviors. This regime lets us evaluate CNNs' ratio estimation ability in bar charts from three perspectives: sensitivity to training-test distribution discrepancies, stability to limited samples, and relative expertise to human observers. After analyzing 16 million trials from 800 CNN models and 6,825 trials from 113 human participants, we arrived at a simple and actionable conclusion: CNNs can outperform humans and their biases simply depend on the training-test distance. We show evidence of this simple, elegant behavior of the machines when they interpret visualization images. osf.io/gfqc3 provides registration, the code for our sampling regime, and experimental results.

In-vitro fertilization (IVF) has become standard practice to address infertility, which affects more than one in ten couples in the US. However, current protocols yield relatively low success rates of about 20% per treatment cycle. A critical but complex and time-consuming step is the grading and selection of embryos for implantation. Although incubators with time-lapse microscopy have enabled computational analysis of embryo development, existing automated approaches either require extensive manual annotations or use opaque deep learning models that are hard for clinicians to validate and trust. We present EmbryoProfiler, a visual analytics system collaboratively developed with embryologists, biologists, and machine learning researchers to support clinicians in visually assessing embryo viability from time-lapse microscopy imagery. Our system incorporates a deep learning pipeline that automatically annotates microscopy images and extracts clinically interpretable features relevant for embryo grading. Our contributions include: (1) a semi-automatic, visualization-based workflow that guides clinicians through fertilization assessment, developmental timing evaluation, morphological inspection, and comparative analysis of embryos; (2) innovative interactive visualizations, such as cell-shape plots, designed to facilitate efficient analysis of morphological and developmental characteristics; and (3) an integrated, explainable machine learning classifier offering transparent, clinically-informed embryo viability scoring to predict live birth outcomes. Quantitative evaluation of our classifier and qualitative case studies conducted with practitioners demonstrate that EmbryoProfiler enables clinicians to make better-informed embryo selection decisions, potentially leading to improved clinical outcomes in IVF treatments.

Connectomics, a subfield of neuroscience, aims to map and analyze synapse-level wiring diagrams of the nervous system. While recent advances in deep learning have accelerated automated neuron and synapse segmentation, reconstructing accurate connectomes still demands extensive human proofreading to correct segmentation errors. We present SynAnno, an interactive tool designed to streamline and enhance the proofreading of synaptic annotations in large-scale connectomics datasets. SynAnno integrates into existing neuroscience workflows by enabling guided, neuron-centric proofreading. To address the challenges posed by the complex spatial branching of neurons, it introduces a structured workflow with an optimized traversal path and a 3D mini-map for tracking progress. In addition, SynAnno incorporates fine-tuned machine learning models to assist with error detection and correction, reducing the manual burden and increasing proofreading efficiency. We evaluate SynAnno through a user and case study involving seven neuroscience experts. Results show that SynAnno significantly accelerates synapse proofreading while reducing cognitive load and annotation errors through structured guidance and visualization support. The source code and interactive demo are available at: https://github.com/PytorchConnectomics/SynAnno.

Connectomics, a subfield of neuroscience, reconstructs structural and functional brain maps at synapse-level resolution. These complex spatial maps consist of tree-like neurons interconnected by synapses. Motif analysis is a widely used method for identifying recurring subgraph patterns in connectomes. These motifs, thus, potentially represent fundamental units of information processing. However, existing computational tools often oversimplify neurons as mere nodes in a graph, disregarding their intricate morphologies. In this paper, we introduce MoMo, a novel interactive visualization framework for analyzing neuron morphology-aware motifs in large connectome graphs. First, we propose an advanced graph data structure that integrates both neuronal morphology and synaptic connectivity. This enables highly efficient, parallel subgraph isomorphism searches, allowing for interactive morphological motif queries. Second, we develop a sketch-based interface that facilitates the intuitive exploration of morphology-based motifs within our new data structure. Users can conduct interactive motif searches on state-of-the-art connectomes and visualize results as interactive 3D renderings. We present a detailed goal and task analysis for motif exploration in connectomes, incorporating neuron morphology. Finally, we evaluate MoMo through case studies with four domain experts, who asses the tool's usefulness and effectiveness in motif exploration, and relevance to real-world neuroscience research. The source code for MoMo is available here.

Dimensionality reduction techniques help analysts make sense of complex, high-dimensional spatial datasets, such as multiplexed tissue imaging, satellite imagery, and astronomical observations, by projecting data attributes into a two-dimensional space. However, these techniques typically abstract away crucial spatial, positional, and morphological contexts, complicating interpretation and limiting insights. To address these limitations, we present SEAL, an interactive visual analytics system designed to bridge the gap between abstract 2D embeddings and their rich spatial imaging context. SEAL introduces a novel hybrid-embedding visualization that preserves image and morphological information while integrating critical high-dimensional feature data. By adapting set visualization methods, SEAL allows analysts to identify, visualize, and compare selections-defined manually or algorithmically-in both the embedding and original spatial views, facilitating a deeper understanding of the spatial arrangement and morphological characteristics of entities of interest. To elucidate differences between selected sets of items, SEAL employs a scalable surrogate model to calculate feature importance scores, identifying the most influential features governing the position of objects within embeddings. These importance scores are visually summarized across selections, with mathematical set operations enabling detailed comparative analyses. We demonstrate SEAL's effectiveness and versatility through three case studies: colorectal cancer tissue analysis with a pharmacologist, melanoma investigation with a cell biologist, and exploration of sky survey data with an astronomer. These studies underscore the importance of integrating image context into embedding spaces when interpreting complex imaging datasets. Implemented as a standalone tool while also integrating seamlessly with computational notebooks, SEAL provides an interactive platform for spatially informed exploration of high-dimensional datasets, significantly enhancing interpretability and insight generation.

Is it better to perform tennis training in a pristine indoor environment or a noisy outdoor one? To model this problem, here we investigate whether shifts in the transition probabilities between the training and testing environments in reinforcement learning problems can lead to better performance under certain conditions. We generate new Markov Decision Processes (MDPs) starting from a given MDP, by adding quantifiable, parametric noise into the transition function. We refer to this process as Noise Injection and the resulting environments as δ-environments. This process allows us to create variations of the same environment with quantitative control over noise serving as a metric of distance between environments. Conventional wisdom suggests that training and testing on the same MDP should yield the best results. In stark contrast, we observe that agents can perform better when trained on the noise-free environment and tested on the noisy δ-environments, compared to training and testing on the same δ-environments. We confirm that this finding extends beyond noise variations: it is possible to showcase the same phenomenon in ATARI game variations including varying Ghost behavior in PacMan, and Paddle behavior in Pong. We demonstrate this intriguing behavior in 60 different variations of ATARI games, including PacMan, Pong, and Breakout. We refer to this phenomenon as the Indoor-Training Effect. Code to reproduce our experiments and to implement Noise Injection.

Learning 4D language fields to enable time-sensitive, open-ended language queries in dynamic scenes is essential for many real-world applications. While LangSplat successfully grounds CLIP features into 3D Gaussian representations, achieving precision and efficiency in 3D static scenes, it lacks the ability to handle dynamic 4D fields as CLIP, designed for static image-text tasks, cannot capture temporal dynamics in videos. Real-world environments are inherently dynamic, with object semantics evolving over time. Building a precise 4D language field necessitates obtaining pixel-aligned, object-wise video features, which current vision models struggle to achieve. To address these challenges, we propose 4D LangSplat, which learns 4D language fields to handle time-agnostic or time-sensitive open-vocabulary queries in dynamic scenes efficiently. 4D LangSplat bypasses learning the language field from vision features and instead learns directly from text generated from object-wise video captions via Multimodal Large Language Models (MLLMs). Specifically, we propose a multimodal object-wise video prompting method, consisting of visual and text prompts that guide MLLMs to generate detailed, temporally consistent, high-quality captions for objects throughout a video. These captions are encoded using a Large Language Model into high-quality sentence embeddings, which then serve as pixel-aligned, object-specific feature supervision, facilitating open-vocabulary text queries through shared embedding spaces. Recognizing that objects in 4D scenes exhibit smooth transitions across states, we further propose a status deformable network to model these continuous changes over time effectively. Our results across multiple benchmarks demonstrate that 4D LangSplat attains precise and efficient results for both timesensitive and time-agnostic open-vocabulary queries.

Prompt learning has proven effective in adapting vision language models for downstream tasks. However, existing methods usually append learnable prompt tokens solely with the category names to obtain textual features, which fails to fully leverage the rich context indicated in the category name. To address this issue, we propose the Tree of Attributes Prompt learning (TAP), which first instructs LLMs to generate a tree of attributes with a “concept - attribute - description” structure for each category, and then learn the hierarchy with vision and text prompt tokens. Unlike existing methods that merely augment category names with a set of unstructured descriptions, our approach essentially distills structured knowledge graphs associated with class names from LLMs. Furthermore, our approach introduces text and vision prompts designed to explicitly learn the corresponding visual attributes, effectively serving as domain experts. Additionally, the general and diverse descriptions generated based on the class names may be wrong or absent in the specific given images. To address this misalignment, we further introduce a vision-conditional pooling module to extract instance-specific text features. Extensive experimental results demonstrate that our approach outperforms state-of-the-art methods on the zero-shot base-to-novel generalization, cross-dataset transfer, as well as few-shot classification across 11 diverse datasets. Code is available at https://github.com/HHenryD/TAP.

Large language models demonstrate impressive performance on downstream tasks, yet requiring extensive resource consumption when fully fine-tuning all parameters. To mitigate this, Parameter Efficient Fine-Tuning (PEFT) strategies, such as LoRA, have been developed. In this paper, we delve into the concept of task-specific directions (TSDs)—critical for transitioning large models from pretrained states to task-specific enhancements in PEFT. We propose a framework to clearly define these directions and explore their properties, and practical utilization challenges. We then introduce a novel approach, LoRA-Dash, which aims to maximize the impact of TSDs during the fine-tuning process, thereby enhancing model performance on targeted tasks. Extensive experiments have conclusively demonstrated the effectiveness of LoRA-Dash, and in-depth analyses further reveal the underlying mechanisms of LoRA-Dash.

Continual Learning (CL) with foundation models has recently emerged as a promising paradigm to exploit abundant knowledge acquired during pre-training for tackling sequential tasks. However, existing prompt-based and Low-Rank Adaptation-based (LoRA-based) methods often require expanding a prompt/LoRA pool or retaining samples of previous tasks, which poses significant scalability challenges as the number of tasks grows. To address these limitations, we propose Scalable Decoupled LoRA (SD-LoRA) for class incremental learning, which continually separates the learning of the magnitude and direction of LoRA components without rehearsal. Our empirical and theoretical analysis reveals that SD-LoRA tends to follow a low-loss trajectory and converges to an overlapping low-loss region for all learned tasks, resulting in an excellent stability-plasticity trade-off. Building upon these insights, we introduce two variants of SD-LoRA with further improved parameter efficiency. All parameters of SD-LoRAs can be end-to-end optimized for CL objectives. Meanwhile, they support efficient inference by allowing direct evaluation with the finally trained model, obviating the need for component selection. Extensive experiments across multiple CL benchmarks and foundation models consistently validate the effectiveness of SD-LoRA. The code is available at https://github.com/WuYichen-97/SD-Lora-CL.

In this paper, we introduce LangSplatV2, which achieves high-dimensional feature splatting at 476.2 FPS and 3D open-vocabulary text querying at 384.6 FPS for high-resolution images, providing a 42 × speedup and a 47 × boost over LangSplat respectively, along with improved query accuracy. LangSplat employs Gaussian Splatting to embed 2D CLIP language features into 3D, significantly enhancing speed and learning a precise 3D language field with SAM semantics. Such advancements in 3D language fields are crucial for applications that require language interaction within complex scenes. However, LangSplat does not yet achieve real- time performance (8.2 FPS), even with advanced A100 GPUs, severely limiting its broader application. In this paper, we first conduct a detailed time analysis of LangSplat, identifying the heavyweight decoder as the primary speed bottleneck. Our solution, LangSplatV2 assumes that each Gaussian acts as a sparse code within a global dictionary, leading to the learning of a 3D sparse coefficient field that entirely eliminates the need for a heavyweight decoder. By leveraging this sparsity, we further propose an efficient sparse coefficient splatting method with CUDA optimization, rendering high-dimensional feature maps at high quality while incurring only the time cost of splatting an ultra-low-dimensional feature. Our experimental results demonstrate that LangSplatV2 not only achieves better or competitive query accuracy but is also significantly faster. Codes and demos are available at our project page: https://langsplat-v2.github.io.

Video storytelling is essential for sports performance analysis and fan engagement, enabling sports professionals and fans to effectively communicate and interpret the spatial and temporal dynamics of gameplay. Traditional methods rely on manual annotation and verbal explanations, placing significant demands on creators for video editing skills and on viewers for cognitive focus. However, these approaches are time-consuming and often struggle to accommodate individual needs. SportsBuddy addresses this gap with an intuitive, interactive video authoring tool. It combines player tracking, embedded interaction design, and timeline visualizations to seamlessly integrate narratives and visual cues within game contexts. This empowers users to effortlessly create context-driven video stories. Since its launch, over 150 sports users, including coaches, athletes, content creators, parents and fans, have utilized SportsBuddy to produce compelling game highlights for diverse use cases. User feedback highlights its accessibility and ease of use, making video storytelling and insight communication more attainable for diverse audiences. Case studies with collegiate teams and sports creators further demonstrate SportsBuddy’s impact on enhancing coaching communication, game analysis, and fan engagement.

Rendering novel, relit views of a human head, given a monocular portrait image as input, is an inherently underconstrained problem. The traditional graphics solution is to explicitly decompose the input image into geometry, material and lighting via differentiable rendering; but this is constrained by the multiple assumptions and approximations of the underlying models and parameterizations of these scene components. We propose 3DPR, an image-based relighting model that leverages generative priors learnt from multi-view One-Light-at-A-Time (OLAT) images captured in a light stage. We introduce a new diverse and large-scale multi-view 4K OLAT dataset of 139 subjects to learn a high-quality prior over the distribution of high-frequency face reflectance. We leverage the latent space of a pre-trained generative head model that provides a rich prior over face geometry learnt from in-the-wild image datasets. The input portrait is first embedded in the latent manifold of such a model through an encoder-based inversion process. Then a novel triplane-based reflectance network trained on our lightstage data is used to synthesize high-fidelity OLAT images to enable image-based relighting. Our reflectance network operates in the latent space of the generative head model, crucially enabling a relatively small number of lightstage images to train the reflectance model. Combining the generated OLATs according to a given HDRI environment maps yields physically accurate environmental relighting results. Through quantitative and qualitative evaluations, we demonstrate that 3DPR outperforms previous methods, particularly in preserving identity and in capturing lighting effects such as specularities, self-shadows, and subsurface scattering.

Recent studies on learning-based sound source localization have primarily focused on localization performance. However, prior work and existing benchmarks often overlook a crucial aspect: cross-modal interaction, which is essential for interactive sound source localization. This interaction is vital for understanding semantically matched or mismatched audio-visual events, such as silent objects or true sound sources among multiple objects. In this work, we comprehensively examine the cross-modal interaction of existing methods, benchmarks, evaluation metrics, and cross-modal understanding tasks. We identify the overlooked points of previous studies and make several contributions to address them. First, we propose a learning framework that incorporates retrieval-based and hand-crafted augmentation techniques, enhancing cross-modal interaction through cross-modal alignment. Second, we introduce new evaluation metrics to accurately and rigorously assess localization methods, focusing on both localization performance and cross-modal interaction. Third, to thoroughly analyze interactive sound source localization, we present a new semi-synthetic benchmark with diverse categorical combinations. Finally, we evaluate both interactive sound source localization and auxiliary cross-modal retrieval tasks, benchmarking competing methods alongside our own. Our new benchmark and evaluation metrics reveal that previous methods struggle with interactive sound source localization tasks, largely due to their limited cross-modal interaction capabilities. Our method, which features enhanced cross-modal alignment, demonstrates superior sound source localization and cross-modal interaction performance. This work provides the most comprehensive analysis of sound source localization to date, with extensive validation of competing methods on both existing and new benchmarks using both new and standard evaluation metrics.

While imaging techniques at macro and mesoscales have garnered substantial attention and resources, microscale Volume Electron Microscopy (vEM) imaging, capable of revealing intricate vascular details, has lacked the necessary benchmarking infrastructure. In this paper, we address a significant gap in this field of neuroimaging by introducing the first-in-class public benchmark, BvEM, designed specifically for cortical blood vessel segmentation in vEM images. Our BvEM benchmark is based on vEM image volumes from three mammals: adult mouse, macaque, and human. We standardized the resolution, addressed imaging variations, and meticulously annotated blood vessels through semi-automatic, manual, and quality control processes, ensuring high-quality 3D segmentation. Furthermore, we developed a zero-shot cortical blood vessel segmentation method named TriSAM, which leverages the powerful segmentation model SAM for 3D segmentation. To extend SAM from 2D to 3D volume segmentation, TriSAM employs a multi-seed tracking framework, leveraging the reliability of certain image planes for tracking while using others to identify potential turning points. This approach effectively achieves long-term 3D blood vessel segmentation without model training or fine-tuning. Experimental results show that TriSAM achieved superior performances on the BvEM benchmark across three species.

Accurate segmentation of anatomical substructures within 3D curvilinear structures in medical imaging remains challenging due to their complex geometry and the scarcity of diverse, large-scale datasets for algorithm development and evaluation. In this paper, we use dendritic spine segmentation as a case study and address these challenges by introducing a novel Frenet-Serret Frame-based Decomposition, which decomposes 3D curvilinear structures into a globally smooth continuous curve that captures the overall shape, and a cylindrical primitive that encodes local geometric properties. This approach leverages Frenet-Serret Frames and arc length parameterization to preserve essential geometric features while reducing representational complexity, facilitating data-efficient learning, improved segmentation accuracy, and generalization on 3D curvilinear structures. To rigorously evaluate our method, we introduce two datasets: CurviSeg, a synthetic dataset for 3D curvilinear structure segmentation that validates our method's key properties, and DenSpineEM, a benchmark for dendritic spine segmentation, which comprises 4,476 manually annotated spines from 70 dendrites across three public electron microscopy datasets, covering multiple brain regions and species. Our experiments on DenSpineEM demonstrate exceptional cross-region and cross-species generalization: models trained on the mouse somatosensory cortex subset achieve 94.43% Dice, maintaining strong performance in zero-shot segmentation on both mouse visual cortex (95.61% Dice) and human frontal lobe (86.63% Dice) subsets. Moreover, we test the generalizability of our method on the IntrA dataset, where it achieves 77.08% Dice (5.29% higher than prior arts) on intracranial aneurysm segmentation from entire artery models. These findings demonstrate the potential of our approach for accurately analyzing complex curvilinear structures across diverse medical imaging fields. Our dataset, code, and models are available at https://github.com/VCG/FFD4DenSpineEM to support future research.

Rib fractures are a common and potentially severe injury that can be challenging and labor-intensive to detect in CT scans. While there have been efforts to address this field, the lack of large-scale annotated datasets and evaluation benchmarks has hindered the development and validation of deep learning algorithms. To address this issue, the RibFrac Challenge was introduced, providing a benchmark dataset of over 5,000 rib fractures from 660 CT scans, with voxel-level instance mask annotations and diagnosis labels for four clinical categories (buckle, nondisplaced, displaced, or segmental). The challenge includes two tracks: a detection (instance segmentation) track evaluated by an FROC-style metric and a classification track evaluated by an F1-style metric. During the MICCAI 2020 challenge period, 243 results were evaluated, and seven teams were invited to participate in the challenge summary. The analysis revealed that several top rib fracture detection solutions achieved performance comparable or even better than human experts. Nevertheless, the current rib fracture classification solutions are hardly clinically applicable, which can be an interesting area in the future. As an active benchmark and research resource, the data and online evaluation of the RibFrac Challenge are available at the challenge website (https://ribfrac.grand-challenge.org/). In addition, we further analyzed the impact of two post-challenge advancements-large-scale pretraining and rib segmentation-based on our internal baseline for rib fracture detection. These findings lay a foundation for future research and development in AI-assisted rib fracture diagnosis.

This article presents a systematic study of the generalization of convolutional neural networks (CNNs) and humans on relational reasoning tasks with bar charts. We first revisit previous experiments on graphical perception and update the benchmark performance of CNNs. We then test the generalization performance of CNNs on a classic relational reasoning task: estimating bar length ratios in a bar chart, by progressively perturbing the standard visualizations. We further conduct a user study to compare the performance of CNNs and humans. Our results show that CNNs outperform humans only when the training and test data have the same visual encodings. Otherwise, they may perform worse. We also find that CNNs are sensitive to perturbations in various visual encodings, regardless of their relevance to the target bars. Yet, humans are mainly influenced by bar lengths. Our study suggests that robust relational reasoning with visualizations is challenging for CNNs. Improving CNNs' generalization performance may require training them to better recognize task-related visual properties.

As basketball's popularity surges, fans often find themselves confused and overwhelmed by the rapid game pace and complexity. Basketball tactics, involving a complex series of actions, require substantial knowledge to be fully understood. This complexity leads to a need for additional information and explanation, which can distract fans from the game. To tackle these challenges, we present Sportify, a Visual Question Answering system that integrates narratives and embedded visualization for demystifying basketball tactical questions, aiding fans in understanding various game aspects. We propose three novel action visualizations (i.e., Pass, Cut, and Screen) to demonstrate critical action sequences. To explain the reasoning and logic behind players' actions, we leverage a large-language model (LLM) to generate narratives. We adopt a storytelling approach for complex scenarios from both first and third-person perspectives, integrating action visualizations. We evaluated Sportify with basketball fans to investigate its impact on understanding of tactics, and how different personal perspectives of narratives impact the understanding of complex tactic with action visualizations. Our evaluation with basketball fans demonstrates Sportify's capability to deepen tactical insights and amplify the viewing experience. Furthermore, third-person narration assists people in getting in-depth game explanations while first-person narration enhances fans' game engagement.

We present Cell2Cell, a novel visual analytics approach for quantifying and visualizing networks of cell-cell interactions in three-dimensional (3D) multi-channel cancerous tissue data. By analyzing cellular interactions, biomedical experts can gain a more accurate understanding of the intricate relationships between cancer and immune cells. Recent methods have focused on inferring interaction based on the proximity of cells in low-resolution 2D multi-channel imaging data. By contrast, we analyze cell interactions by quantifying the presence and levels of specific proteins within a tissue sample (protein expressions) extracted from high-resolution 3D multi-channel volume data. Such analyses have a strong exploratory nature and require a tight integration of domain experts in the analysis loop to leverage their deep knowledge. We propose two complementary semi-automated approaches to cope with the increasing size and complexity of the data interactively: On the one hand, we interpret cell-to-cell interactions as edges in a cell graph and analyze the image signal (protein expressions) along those edges, using spatial as well as abstract visualizations. Complementary, we propose a cell-centered approach, enabling scientists to visually analyze polarized distributions of proteins in three dimensions, which also captures neighboring cells with biochemical and cell biological consequences. We evaluate our application in three case studies, where biologists and medical experts use Cell2Cell to investigate tumor micro-environments to identify and quantify T-cell activation in human tissue data. We confirmed that our tool can fully solve the use cases and enables a streamlined and detailed analysis of cell-cell interactions.

In our current visual-centric digital age, the capability to interpret, understand, and produce visual representations of data —termed visualization literacy— is paramount. However, not everyone is adept at navigating this visual terrain. This paper explores the barriers that individuals who misread a visualization encounter, aiming to understand their specific mental gaps.

Multi-task learning has become increasingly popular in the machine learning field, but its practicality is hindered by the need for large, labeled datasets. Most multi-task learning methods depend on fully labeled datasets wherein each input example is accompanied by ground-truth labels for all target tasks. Unfortunately, curating such datasets can be prohibitively expensive and impractical, especially for dense prediction tasks which require per-pixel labels for each image. With this in mind, we propose Joint-Task Regularization (JTR), an intuitive technique which leverages cross-task relations to simultaneously regularize all tasks in a single joint-task latent space to improve learning when data is not fully labeled for all tasks. JTR stands out from existing approaches in that it regularizes all tasks jointly rather than separately in pairs-therefore, it achieves linear complexity relative to the number of tasks while previous methods scale quadratically. To demonstrate the validity of our approach, we extensively benchmark our method across a wide variety of partially labeled scenarios based on NYU-v2, Cityscapes, and Taskonomy.

Humans live in a 3D world and commonly use natural language to interact with a 3D scene. Modeling a 3D language field to support open-ended language queries in 3D has gained increasing attention recently. This paper introduces LangSplat, which constructs a 3D language field that enables precise and efficient open-vocabulary querying within 3D spaces. Unlike existing methods that ground CLIP language embeddings in a NeRF model, LangSplat advances the field by utilizing a collection of 3D Gaussians, each encoding language features distilled from CLIP, to represent the language field. By employing a tile-based splatting technique for rendering language features, we circumvent the costly rendering process inherent in NeRF. Instead of directly learning CLIP embeddings, LangSplat first trains a scene-wise language autoencoder and then learns language features on the scene-specific latent space, thereby alleviating substantial memory demands imposed by explicit modeling. Existing methods struggle with imprecise and vague 3D language fields, which fail to discern clear boundaries between objects. We delve into this issue and propose to learn hierarchical semantics using SAM, thereby eliminating the need for extensively querying the language field across various scales and the regularization of DINO features. Extensive experimental results show that LangSplat significantly outperforms the previous state-of-the-art method LERF by a large margin. Notably, LangSplat is extremely efficient, achieving a 199 x speedup compared to LERF at the resolution of 1440 x 1080. We strongly recommend readers to check out our video results at https://langsplat.github.io/

Social relation reasoning aims to identify relation categories such as friends, spouses, and colleagues from images. While current methods adopt the paradigm of training a dedicated network end-to-end using labeled image data, they are limited in terms of generalizability and interpretability. To address these issues, we first present a simple yet well-crafted framework named SocialGPT, which combines the perception capability of Vision Foundation Models (VFMs) and the reasoning capability of Large Language Models (LLMs) within a modular framework, providing a strong baseline for social relation recognition. Specifically, we instruct VFMs to translate image content into a textual social story, and then utilize LLMs for text-based reasoning. SocialGPT introduces systematic design principles to adapt VFMs and LLMs separately and bridge their gaps. Without additional model training, it achieves competitive zero-shot results on two databases while offering interpretable answers, as LLMs can generate language-based explanations for the decisions. The manual prompt design process for LLMs at the reasoning phase is tedious and an automated prompt optimization method is desired. As we essentially convert a visual classification task into a generative task of LLMs, automatic prompt optimization encounters a unique long prompt optimization issue. To address this issue, we further propose the Greedy Segment Prompt Optimization (GSPO), which performs a greedy search by utilizing gradient information at the segment level. Experimental results show that GSPO significantly improves performance, and our method also generalizes to different image styles. The code is available at https://github.com/Mengzibin/SocialGPT.

We characterized the generalization capabilities of deep neural network encoding models when predicting neuronal responses from the visual cortex to flashed images. We collected MacaqueITBench, a large-scale dataset of neuronal population responses from the macaque inferior temporal (IT) cortex to over $300,000$ images, comprising $8,233$ unique natural images presented to seven monkeys over $109$ sessions. Using MacaqueITBench, we investigated the impact of distribution shifts on models predicting neuronal activity by dividing the images into Out-Of-Distribution (OOD) train and test splits. The OOD splits included variations in image contrast, hue, intensity, temperature, and saturation. Compared to the performance on in-distribution test images---the conventional way in which these models have been evaluated---models performed worse at predicting neuronal responses to out-of-distribution images, retaining as little as $20\\%$ of the performance on in-distribution test images. Additionally, the relative ranking of different models in terms of their ability to predict neuronal responses changed drastically across OOD shifts. The generalization performance under OOD shifts can be well accounted by a simple image similarity metric---the cosine distance between image representations extracted from a pre-trained object recognition model is a strong predictor of neuronal predictivity under different distribution shifts. The dataset of images, neuronal firing rate recordings, and computational benchmarks are hosted publicly at: https://github.com/Spandan-Madan/benchmarking_ood_generalization_visual_cortex.

The rapid development of generative AI (GenAI) models in computer vision necessitates effective evaluation methods to ensure their quality and fairness. Existing tools primarily focus on dataset quality assurance and model explainability, leaving a significant gap in GenAI output evaluation during model development. Current practices often depend on developers’ subjective visual assessments, which may lack scalability and generalizability. This paper bridges this gap by conducting a formative study with GenAI model developers in an industrial setting. Our findings led to the development of GenLens, a visual analytic interface designed for the systematic evaluation of GenAI model outputs during the early stages of model development. GenLens offers a quantifiable approach for overviewing and annotating failure cases, customizing issue tags and classifications, and aggregating annotations from multiple users to enhance collaboration. A user study with model developers reveals that GenLens effectively enhances their workflow, evidenced by high satisfaction rates and a strong intent to integrate it into their practices. This research underscores the importance of robust early-stage evaluation tools in GenAI development, contributing to the advancement of fair and high-quality GenAI models.

Over the past century, multichannel fluorescence imaging has been pivotal in myriad scientific breakthroughs by enabling the spatial visualization of proteins within a biological sample. With the shift to digital methods and visualization software, experts can now flexibly pseudocolor and combine image channels, each corresponding to a different protein, to explore their spatial relationships. We thus propose psudo, an interactive system that allows users to create optimal color palettes for multichannel spatial data. In psudo, a novel optimization method generates palettes that maximize the perceptual differences between channels while mitigating confusing color blending in overlapping channels. We integrate this method into a system that allows users to explore multi-channel image data and compare and evaluate color palettes for their data. An interactive lensing approach provides on-demand feedback on channel overlap and a color confusion metric while giving context to the underlying channel values. Color palettes can be applied globally or, using the lens, to local regions of interest. We evaluate our palette optimization approach using three graphical perception tasks in a crowdsourced user study with 150 participants, showing that users are more accurate at discerning and comparing the underlying data using our approach. Additionally, we showcase psudo in a case study exploring the complex immune responses in cancer tissue data with a biologist.

Abstract Portrait viewpoint and illumination editing is an important problem with several applications in VR/AR, movies, and photography. Comprehensive knowledge of geometry and illumination is critical for obtaining photorealistic results. Current methods are unable to explicitly model in 3D while handling both viewpoint and illumination editing from a single image. In this paper, we propose VoRF, a novel approach that can take even a single portrait image as input and relight human heads under novel illuminations that can be viewed from arbitrary viewpoints. VoRF represents a human head as a continuous volumetric field and learns a prior model of human heads using a coordinate-based MLP with individual latent spaces for identity and illumination. The prior model is learned in an auto-decoder manner over a diverse class of head shapes and appearances, allowing VoRF to generalize to novel test identities from a single input image. Additionally, VoRF has a reflectance MLP that uses the intermediate features of the prior model for rendering One-Light-at-A-Time (OLAT) images under novel views. We synthesize novel illuminations by combining these OLAT images with target environment maps. Qualitative and quantitative evaluations demonstrate the effectiveness of VoRF for relighting and novel view synthesis, even when applied to unseen subjects under uncontrolled illumination. This work is an extension of Rao et al. (VoRF: Volumetric Relightable Faces 2022). We provide extensive evaluation and ablative studies of our model and also provide an application, where any face can be relighted using textual input.

Automatic rib labeling and anatomical centerline extraction are common prerequisites for various clinical applications. Prior studies either use in-house datasets that are inaccessible to communities, or focus on rib segmentation that neglects the clinical significance of rib labeling. To address these issues, we extend our prior dataset (RibSeg) on the binary rib segmentation task to a comprehensive benchmark, named RibSeg v2, with 660 CT scans (15,466 individual ribs in total) and annotations manually inspected by experts for rib labeling and anatomical centerline extraction. Based on the RibSeg v2, we develop a pipeline including deep learning-based methods for rib labeling, and a skeletonization-based method for centerline extraction. To improve computational efficiency, we propose a sparse point cloud representation of CT scans and compare it with standard dense voxel grids. Moreover, we design and analyze evaluation metrics to address the key challenges of each task. Our dataset, code, and model are available online to facilitate open research at https://github.com/M3DV/RibSeg.

The size of image volumes in connectomics studies now reaches terabyte and often petabyte scales with a great diversity of appearance due to different sample preparation procedures. However, manual annotation of neuronal structures (e.g., synapses) in these huge image volumes is time-consuming, leading to limited labeled training data often smaller than 0.001% of the large-scale image volumes in application. Methods that can utilize in-domain labeled data and generalize to out-of-domain unlabeled data are in urgent need. Although many domain adaptation approaches are proposed to address such issues in the natural image domain, few of them have been evaluated on connectomics data due to a lack of domain adaptation benchmarks. Therefore, to enable developments of domain adaptive synapse detection methods for large-scale connectomics applications, we annotated 14 image volumes from a biologically diverse set of Megaphragma viggianii brain regions originating from three different whole-brain datasets and organized the WASPSYN challenge at ISBI 2023. The annotations include coordinates of pre-synapses and post-synapses in the 3D space, together with their one-to-many connectivity information. This paper describes the dataset, the tasks, the proposed baseline, the evaluation method, and the results of the challenge. Limitations of the challenge and the impact on neuroscience research are also discussed. The challenge is and will continue to be available at https://codalab.lisn.upsaclay.fr/competitions/9169. Successful algorithms that emerge from our challenge may potentially revolutionize real-world connectomics research and further the cause that aims to unravel the complexity of brain structure and function.

We present VoxAR, a method to facilitate an effective visualization of volume-rendered objects in optical see-through head-mounted displays (OST-HMDs). The potential of augmented reality (AR) to integrate digital information into the physical world provides new opportunities for visualizing and interpreting scientific data. However, a limitation of OST-HMD technology is that rendered pixels of a virtual object can interfere with the colors of the real-world, making it challenging to perceive the augmented virtual information accurately. We address this challenge in a two-step approach. First, VoxAR determines an appropriate placement of the volume-rendered object in the real-world scene by evaluating a set of spatial and environmental objectives, managed as user-selected preferences and pre-defined constraints. We achieve a real-time solution by implementing the objectives using a GPU shader language. Next, VoxAR adjusts the colors of the input transfer function (TF) based on the real-world placement region. Specifically, we introduce a novel optimization method that adjusts the TF colors such that the resulting volume-rendered pixels are discernible against the background and the TF maintains the perceptual mapping between the colors and data intensity values. Finally, we present an assessment of our approach through objective evaluations and subjective user studies.

Labels are widely used in augmented reality (AR) to display digital information. Ensuring the readability of AR labels requires placing them in an occlusion-free manner while keeping visual links legible, especially when multiple labels exist in the scene. Although existing optimization-based methods, such as force-based methods, are effective in managing AR labels in static scenarios, they often struggle in dynamic scenarios with constantly moving objects. This is due to their focus on generating layouts optimal for the current moment, neglecting future moments and leading to sub-optimal or unstable layouts over time. In this work, we present RL-LABEL, a deep reinforcement learning-based method intended for managing the placement of AR labels in scenarios involving moving objects. RL-LABEL considers both the current and predicted future states of objects and labels, such as positions and velocities, as well as the user's viewpoint, to make informed decisions about label placement. It balances the trade-offs between immediate and long-term objectives. We tested RL-LABEL in simulated AR scenarios on two real-world datasets, showing that it effectively learns the decision-making process for long-term optimization, outperforming two baselines (i.e., no view management and a force-based method) by minimizing label occlusions, line intersections, and label movement distance. Additionally, a user study involving 18 participants indicates that, within our simulated environment, RL-LABEL excels over the baselines in aiding users to identify, compare, and summarize data on labels in dynamic scenes.

Existing model evaluation tools mainly focus on evaluating classification models, leaving a gap in evaluating more complex models, such as object detection. In this paper, we develop an open-source visual analysis tool, Uni-Evaluator, to support a unified model evaluation for classification, object detection, and instance segmentation in computer vision. The key idea behind our method is to formulate both discrete and continuous predictions in different tasks as unified probability distributions. Based on these distributions, we develop 1) a matrix-based visualization to provide an overview of model performance; 2) a table visualization to identify the problematic data subsets where the model performs poorly; 3) a grid visualization to display the samples of interest. These visualizations work together to facilitate the model evaluation from a global overview to individual samples. Two case studies demonstrate the effectiveness of Uni-Evaluator in evaluating model performance and making informed improvements.

Trust is an essential aspect of data visualization, as it plays a crucial role in the interpretation and decision-making processes of users. While research in social sciences outlines the multi-dimensional factors that can play a role in trust formation, most data visualization trust researchers employ a single-item scale to measure trust. We address this gap by proposing a comprehensive, multidimensional conceptualization and operationalization of trust in visualization. We do this by applying general theories of trust from social sciences, as well as synthesizing and extending earlier work and factors identified by studies in the visualization field. We apply a two-dimensional approach to trust in visualization, to distinguish between cognitive and affective elements, as well as between visualization and data-specific trust antecedents. We use our framework to design and run a large crowd-sourced study to quantify the role of visual complexity in establishing trust in science visualizations. Our study provides empirical evidence for several aspects of our proposed theoretical framework, most notably the impact of cognition, affective responses, and individual differences when establishing trust in visualizations.

We present a hybrid multi-volume rendering approach based on a novel Residency Octree that combines the advantages of out-of-core volume rendering using page tables with those of standard octrees. Octree approaches work by performing hierarchical tree traversal. However, in octree volume rendering, tree traversal and the selection of data resolution are intrinsically coupled. This makes fine-grained empty-space skipping costly. Page tables, on the other hand, allow access to any cached brick from any resolution. However, they do not offer a clear and efficient strategy for substituting missing high-resolution data with lower-resolution data. We enable flexible mixed-resolution out-of-core multi-volume rendering by decoupling the cache residency of multi-resolution data from a resolution-independent spatial subdivision determined by the tree. Instead of one-to-one node-to-brick correspondences, each residency octree node is mapped to a set of bricks from different resolution levels. This makes it possible to efficiently and adaptively choose and mix resolutions, adapt sampling rates, and compensate for cache misses. At the same time, residency octrees support fine-grained empty-space skipping, independent of the data subdivision used for caching. Finally, to facilitate collaboration and outreach, and to eliminate local data storage, our implementation is a web-based, pure client-side renderer using WebGPU and WebAssembly. Our method is faster than prior approaches and efficient for many data channels with a flexible and adaptive choice of data resolution.

Data transformation is an essential step in data science. While experts primarily use programming to transform their data, there is an increasing need to support non-programmers with user interface-based tools. With the rapid development in interaction techniques and computing environments, we report our empirical findings about the effects of interaction techniques and environments on performing data transformation tasks. Specifically, we studied the potential benefits of direct interaction and virtual reality (VR) for data transformation. We compared gesture interaction versus a standard WIMP user interface, each on the desktop and in VR. With the tested data and tasks, we found time performance was similar between desktop and VR. Meanwhile, VR demonstrates preliminary evidence to better support provenance and sense-making throughout the data transformation process. Our exploration of performing data transformation in VR also provides initial affirmation for enabling an iterative and fully immersive data science workflow.

Badminton is a fast-paced sport that requires a strategic combination of spatial, temporal, and technical tactics. To gain a competitive edge at high-level competitions, badminton professionals frequently analyze match videos to gain insights and develop game strategies. However, the current process for analyzing matches is time-consuming and relies heavily on manual note-taking, due to the lack of automatic data collection and appropriate visualization tools. As a result, there is a gap in effectively analyzing matches and communicating insights among badminton coaches and players. This work proposes an end-to-end immersive match analysis pipeline designed in close collaboration with badminton professionals, including Olympic and national coaches and players. We present VIRD, a VR Bird (i.e., shuttle) immersive analysis tool, that supports interactive badminton game analysis in an immersive environment based on 3D reconstructed game views of the match video. We propose a top-down analytic workflow that allows users to seamlessly move from a high-level match overview to a detailed game view of individual rallies and shots, using situated 3D visualizations and video. We collect 3D spatial and dynamic shot data and player poses with computer vision models and visualize them in VR. Through immersive visualizations, coaches can interactively analyze situated spatial data (player positions, poses, and shot trajectories) with flexible viewpoints while navigating between shots and rallies effectively with embodied interaction. We evaluated the usefulness of VIRD with Olympic and national-level coaches and players in real matches. Results show that immersive analytics supports effective badminton match analysis with reduced context-switching costs and enhances spatial understanding with a high sense of presence.

Recent advances in high-resolution connectomics provide researchers with access to accurate petascale reconstructions of neuronal circuits and brain networks for the first time. Neuroscientists are analyzing these networks to better understand information processing in the brain. In particular, scientists are interested in identifying specific small network motifs, i.e., repeating subgraphs of the larger brain network that are believed to be neuronal building blocks. Although such motifs are typically small (e.g., 2 - 6 neurons), the vast data sizes and intricate data complexity present significant challenges to the search and analysis process. To analyze these motifs, it is crucial to review instances of a motif in the brain network and then map the graph structure to detailed 3D reconstructions of the involved neurons and synapses. We present Vimo, an interactive visual approach to analyze neuronal motifs and motif chains in large brain networks. Experts can sketch network motifs intuitively in a visual interface and specify structural properties of the involved neurons and synapses to query large connectomics datasets. Motif instances (MIs) can be explored in high-resolution 3D renderings. To simplify the analysis of MIs, we designed a continuous focus&context metaphor inspired by visual abstractions. This allows users to transition from a highly-detailed rendering of the anatomical structure to views that emphasize the underlying motif structure and synaptic connectivity. Furthermore, Vimo supports the identification of motif chains where a motif is used repeatedly (e.g., 2 - 4 times) to form a larger network structure. We evaluate Vimo in a user study and an in-depth case study with seven domain experts on motifs in a large connectome of the fruit fly, including more than 21,000 neurons and 20 million synapses. We find that Vimo enables hypothesis generation and confirmation through fast analysis iterations and connectivity highlighting.

Volumetric data abounds in medical imaging and other fields. With the improved imaging quality and the increased resolution, volumetric datasets are getting so large that the existing tools have become inadequate for processing and analyzing the data. Here we consider the problem of computing tetrahedral meshes to represent large volumetric datasets with labeled multiple materials, which are often encountered in medical imaging or microscopy optical slice tomography. Such tetrahedral meshes are a more compact and expressive geometric representation so are in demand for efficient visualization and simulation of the data, which are impossible if the original large volumetric data are used directly due to the large memory requirement. Existing methods for meshing volumetric data are not scalable for handling large datasets due to their sheer demand on excessively large run-time memory or failure to produce a tet-mesh that preserves the multi-material structure of the original volumetric data. In this article we propose a novel approach, called Marching Windows, that uses a moving window and a disk-swap strategy to reduce the run-time memory footprint, devise a new scheme that guarantees to preserve the topological structure of the original dataset, and adopt an error-guided optimization technique to improve both geometric approximation error and mesh quality. Extensive experiments show that our method is capable of processing very large volumetric datasets beyond the capability of the existing methods and producing tetrahedral meshes of high quality.

We present iBall, a basketball video-watching system that leverages gaze-moderated embedded visualizations to facilitate game understanding and engagement of casual fans. Video broadcasting and online video platforms make watching basketball games increasingly accessible. Yet, for new or casual fans, watching basketball videos is often confusing due to their limited basketball knowledge and the lack of accessible, on-demand information to resolve their confusion. To assist casual fans in watching basketball videos, we compared the game-watching behaviors of casual and die-hard fans in a formative study and developed iBall based on the findings. iBall embeds visualizations into basketball videos using a computer vision pipeline, and automatically adapts the visualizations based on the game context and users’ gaze, helping casual fans appreciate basketball games without being overwhelmed. We confirmed the usefulness, usability, and engagement of iBall in a study with 16 casual fans, and further collected feedback from 8 die-hard fans.

When capturing and storing images, devices inevitably introduce noise. Reducing this noise is a critical task called image denoising. Deep learning has become the de facto method for image denoising, especially with the emergence of Transformer-based models that have achieved notable state-of-the-art results on various image tasks. However, deep learning-based methods often suffer from a lack of generalization ability. For example, deep models trained on Gaussian noise may perform poorly when tested on other noise distributions. To address this issue, we present a novel approach to enhance the generalization performance of denoising networks, known as masked training. Our method involves masking random pixels of the input image and reconstructing the missing information during training. We also mask out the features in the self-attention layers to avoid the impact of training-testing inconsistency. Our approach exhibits better generalization ability than other deep learning models and is directly applicable to real-world scenarios. Additionally, our interpretability analysis demonstrates the superiority of our method.

Recent deep learning methods have achieved promising results in image shadow removal. However, their restored images still suffer from unsatisfactory boundary artifacts, due to the lack of degradation prior embedding and the deficiency in modeling capacity. Our work addresses these issues by proposing a unified diffusion framework that integrates both the image and degradation priors for highly effective shadow removal. In detail, we first propose a shadow degradation model, which inspires us to build a novel unrolling diffusion model, dubbed ShandowDiffusion. It remarkably improves the model's capacity in shadow removal via progressively refining the desired output with both degradation prior and diffusive generative prior, which by nature can serve as a new strong baseline for image restoration. Furthermore, ShadowDiffusion progressively refines the estimated shadow mask as an auxiliary task of the diffusion generator, which leads to more accurate and robust shadow-free image generation. We conduct extensive experiments on three popular public datasets, including ISTD, ISTD+, and SRD, to validate our method's effectiveness. Compared to the state-of-the-art methods, our model achieves a significant improvement in terms of PSNR, increasing from 31.69dB to 34. 73dB over SRD dataset.11https://github.com/GuoLanqing/ShadowDiffusion

The mechanism of existing style transfer algorithms is by minimizing a hybrid loss function to push the generated image toward high similarities in both content and style. However, this type of approach cannot guarantee visual fidelity, i.e., the generated artworks should be indistinguishable from real ones. In this paper, we devise a new style transfer framework called QunntArt for high visual-fidelity stylization. QuantArt pushes the latent representation of the generated artwork toward the centroids of the real artwork distribution with vector quantization. By fusing the quantized and continuous latent representations, QuantArt allows flexible control over the generated artworks in terms of content preservation, style similarity, and visual fidelity. Experiments on various style transfer settings show that our QuantArt framework achieves significantly higher visual fidelity compared with the existing style transfer methods.

There has been a growing interest in developing multimodal machine translation (MMT) systems that enhance neural machine translation (NMT) with visual knowledge. This problem setup involves using images as auxiliary information during training, and more recently, eliminating their use during inference. Towards this end, previous works face a challenge in training powerful MMT models from scratch due to the scarcity of annotated multilingual vision-language data, especially for low-resource languages. Simultaneously, there has been an influx of multilingual pretrained models for NMT and multimodal pre-trained models for vision-language tasks, primarily in English, which have shown exceptional generalisation ability. However, these are not directly applicable to MMT since they do not provide aligned multimodal multilingual features for generative tasks. To alleviate this issue, instead of designing complex modules for MMT, we propose CLIPTrans, which simply adapts the independently pre-trained multimodal M-CLIP and the multilingual mBART. In order to align their embedding spaces, mBART is conditioned on the M-CLIP features by a prefix sequence generated through a lightweight mapping network. We train this in a two-stage pipeline which warms up the model with image captioning before the actual translation task. Through experiments, we demonstrate the merits of this framework and consequently push forward the state-of-the-art across standard benchmarks by an average of +2.67 BLEU. The code can be found at www.github.com/devaansh100/CLIPTrans.

Humans can easily perceive the direction of sound sources in a visual scene, termed sound source localization. Recent studies on learning-based sound source localization have mainly explored the problem from a localization perspective. However, prior arts and existing benchmarks do not account for a more important aspect of the problem, cross-modal semantic understanding, which is essential for genuine sound source localization. Cross-modal semantic understanding is important in understanding semantically mismatched audio-visual events, e.g., silent objects, or off-screen sounds. To account for this, we propose a cross-modal alignment task as a joint task with sound source localization to better learn the interaction between audio and visual modalities. Thereby, we achieve high localization performance with strong cross-modal semantic understanding. Our method outperforms the state-of-the-art approaches in both sound source localization and cross-modal retrieval. Our work suggests that jointly tackling both tasks is necessary to conquer genuine sound source localization.

We introduce Inference-Time Intervention (ITI), a technique designed to enhance the "truthfulness" of large language models (LLMs). ITI operates by shifting model activations during inference, following a learned set of directions across a limited number of attention heads. This intervention significantly improves the performance of LLaMA models on the TruthfulQA benchmark. On an instruction-finetuned LLaMA called Alpaca, ITI improves its truthfulness from $32.5\%$ to $65.1\%$. We identify a tradeoff between truthfulness and helpfulness and demonstrate how to balance it by tuning the intervention strength. ITI is minimally invasive and computationally inexpensive. Moreover, the technique is data efficient: while approaches like RLHF require extensive annotations, ITI locates truthful directions using only few hundred examples. Our findings suggest that LLMs may have an internal representation of the likelihood of something being true, even as they produce falsehoods on the surface.

Unpaired image-to-image translation (UNIT) aims to map images between two visual domains without paired training data. However, given a UNIT model trained on certain domains, it is difficult for current methods to incorporate new domains because they often need to train the full model on both existing and new domains. To address this problem, we propose a new domain-scalable UNIT method, termed as latent space anchoring, which can be efficiently extended to new visual domains and does not need to fine-tune encoders and decoders of existing domains. Our method anchors images of different domains to the same latent space of frozen GANs by learning lightweight encoder and regressor models to reconstruct single-domain images. In the inference phase, the learned encoders and decoders of different domains can be arbitrarily combined to translate images between any two domains without fine-tuning. Experiments on various datasets show that the proposed method achieves superior performance on both standard and domain-scalable UNIT tasks in comparison with the state-of-the-art methods.

3D instance segmentation for unlabeled imaging modalities is a challenging but essential task as collecting expert annotation can be expensive and time-consuming. Existing works segment a new modality by either deploying pre-trained models optimized on diverse training data or sequentially conducting image translation and segmentation with two relatively independent networks. In this work, we propose a novel Cyclic Segmentation Generative Adversarial Network (CySGAN) that conducts image translation and instance segmentation simultaneously using a unified network with weight sharing. Since the image translation layer can be removed at inference time, our proposed model does not introduce additional computational cost upon a standard segmentation model. For optimizing CySGAN, besides the CycleGAN losses for image translation and supervised losses for the annotated source domain, we also utilize self-supervised and segmentation-based adversarial objectives to enhance the model performance by leveraging unlabeled target domain images. We benchmark our approach on the task of 3D neuronal nuclei segmentation with annotated electron microscopy (EM) images and unlabeled expansion microscopy (ExM) data. The proposed CySGAN outperforms pre-trained generalist models, feature-level domain adaptation models, and the baselines that conduct image translation and segmentation sequentially. Our implementation and the newly collected, densely annotated ExM zebrafish brain nuclei dataset, named NucExM, are publicly available at https://connectomics-bazaar.github.io/proj/CySGAN/index.html.

In this paper, we present the results of the MitoEM challenge on mitochondria 3D instance segmentation from electron microscopy images, organized in conjunction with the IEEE-ISBI 2021 conference. Our benchmark dataset consists of two large-scale 3D volumes, one from human and one from rat cortex tissue, which are 1,986 times larger than previously used datasets. At the time of paper submission, 257 participants had registered for the challenge, 14 teams had submitted their results, and six teams participated in the challenge workshop. Here, we present eight top-performing approaches from the challenge participants, along with our own baseline strategies. Posterior to the challenge, annotation errors in the ground truth were corrected without altering the final ranking. Additionally, we present a retrospective evaluation of the scoring system which revealed that: 1) challenge metric was permissive with the false positive predictions; and 2) size-based grouping of instances did not correctly categorize mitochondria of interest. Thus, we propose a new scoring system that better reflects the correctness of the segmentation results. Although several of the top methods are compared favorably to our own baselines, substantial errors remain unsolved for mitochondria with challenging morphologies. Thus, the challenge remains open for submission and automatic evaluation, with all volumes available for download.

Augmented sports videos, which combine visualizations and video effects to present data in actual scenes, can communicate insights engagingly and thus have been increasingly popular for sports enthusiasts around the world. Yet, creating augmented sports videos remains a challenging task, requiring considerable time and video editing skills. On the other hand, sports insights are often communicated using natural language, such as in commentaries, oral presentations, and articles, but usually lack visual cues. Thus, this work aims to facilitate the creation of augmented sports videos by enabling analysts to directly create visualizations embedded in videos using insights expressed in natural language. To achieve this goal, we propose a three-step approach - 1) detecting visualizable entities in the text, 2) mapping these entities into visualizations, and 3) scheduling these visualizations to play with the video - and analyzed 155 sports video clips and the accompanying commentaries for accomplishing these steps. Informed by our analysis, we have designed and implemented Sporthesia, a proof-of-concept system that takes racket-based sports videos and textual commentaries as the input and outputs augmented videos. We demonstrate Sporthesia's applicability in two exemplar scenarios, i.e., authoring augmented sports videos using text and augmenting historical sports videos based on auditory comments. A technical evaluation shows that Sporthesia achieves high accuracy (F1-score of 0.9) in detecting visualizable entities in the text. An expert evaluation with eight sports analysts suggests high utility, effectiveness, and satisfaction with our language-driven authoring method and provides insights for future improvement and opportunities.

Sports game data is becoming increasingly complex, often consisting of multivariate data such as player performance stats, historical team records, and athletes' positional tracking information. While numerous visual analytics systems have been developed for sports analysts to derive insights, few tools target fans to improve their understanding and engagement of sports data during live games. By presenting extra data in the actual game views, embedded visualization has the potential to enhance fans' game-viewing experience. However, little is known about how to design such kinds of visualizations embedded into live games. In this work, we present a user-centered design study of developing interactive embedded visualizations for basketball fans to improve their live game-watching experiences. We first conducted a formative study to characterize basketball fans' in-game analysis behaviors and tasks. Based on our findings, we propose a design framework to inform the design of embedded visualizations based on specific data-seeking contexts. Following the design framework, we present five novel embedded visualization designs targeting five representative contexts identified by the fans, including shooting, offense, defense, player evaluation, and team comparison. We then developed Omnioculars, an interactive basketball game-viewing prototype that features the proposed embedded visualizations for fans' in-game data analysis. We evaluated Omnioculars in a simulated basketball game with basketball fans. The study results suggest that our design supports personalized in-game data analysis and enhances game understanding and engagement.

Augmented Reality (AR) embeds digital information into objects of the physical world. Data can be shown in-situ, thereby enabling real-time visual comparisons and object search in real-life user tasks, such as comparing products and looking up scores in a sports game. While there have been studies on designing AR interfaces for situated information retrieval, there has only been limited research on AR object labeling for visual search tasks in the spatial environment. In this article, we identify and categorize different design aspects in AR label design and report on a formal user study on labels for out-of-view objects to support visual search tasks in AR. We design three visualization techniques for out-of-view object labeling in AR, which respectively encode the relative physical position (height-encoded), the rotational direction (angle-encoded), and the label values (value-encoded) of the objects. We further implement two traditional in-view object labeling techniques, where labels are placed either next to the respective objects (situated) or at the edge of the AR FoV (boundary). We evaluate these five different label conditions in three visual search tasks for static objects. Our study shows that out-of-view object labels are beneficial when searching for objects outside the FoV, spatial orientation, and when comparing multiple spatially sparse objects. Angle-encoded labels with directional cues of the surrounding objects have the overall best performance with the highest user satisfaction. We discuss the implications of our findings for future immersive AR interface design.

State-of-the-art neural language models can now be used to solve ad-hoc language tasks through zero-shot prompting without the need for supervised training. This approach has gained popularity in recent years, and researchers have demonstrated prompts that achieve strong accuracy on specific NLP tasks. However, finding a prompt for new tasks requires experimentation. Different prompt templates with different wording choices lead to significant accuracy differences. PromptIDE allows users to experiment with prompt variations, visualize prompt performance, and iteratively optimize prompts. We developed a workflow that allows users to first focus on model feedback using small data before moving on to a large data regime that allows empirical grounding of promising prompts using quantitative measures of the task. The tool then allows easy deployment of the newly created ad-hoc models. We demonstrate the utility of PromptIDE (demo: http://prompt.vizhub.ai) and our workflow using several real-world use cases.

New highly-multiplexed imaging technologies have enabled the study of tissues in unprecedented detail. These methods are increasingly being applied to understand how cancer cells and immune response change during tumor development, progression, and metastasis, as well as following treatment. Yet, existing analysis approaches focus on investigating small tissue samples on a per-cell basis, not taking into account the spatial proximity of cells, which indicates cell-cell interaction and specific biological processes in the larger cancer microenvironment. We present Visinity, a scalable visual analytics system to analyze cell interaction patterns across cohorts of whole-slide multiplexed tissue images. Our approach is based on a fast regional neighborhood computation, leveraging unsupervised learning to quantify, compare, and group cells by their surrounding cellular neighborhood. These neighborhoods can be visually analyzed in an exploratory and confirmatory workflow. Users can explore spatial patterns present across tissues through a scalable image viewer and coordinated views highlighting the neighborhood composition and spatial arrangements of cells. To verify or refine existing hypotheses, users can query for specific patterns to determine their presence and statistical significance. Findings can be interactively annotated, ranked, and compared in the form of small multiples. In two case studies with biomedical experts, we demonstrate that Visinity can identify common biological processes within a human tonsil and uncover novel white-blood cell networks and immune-tumor interactions.

Matrix visualizations are widely used to display large-scale network, tabular, set, or sequential data. They typically only encode a single value per cell, e.g., through color. However, this can greatly limit the visualizations’ utility when exploring multivariate data, where each cell represents a data point with multiple values (referred to as details). Three well-established interaction approaches can be applicable in multivariate matrix visualizations (or MMV): focus+context, pan&zoom, and overview+detail. However, there is little empirical knowledge of how these approaches compare in exploring MMV. We report on two studies comparing them for locating, searching, and contextualizing details in MMV. We first compared four focus+context techniques and found that the fisheye lens overall outperformed the others. We then compared the fisheye lens, to pan&zoom and overview+detail. We found that pan&zoom was faster in locating and searching details, and as good as overview+detail in contextualizing details.

Many video understanding tasks require analyzing multishot videos, but existing datasets for video object segmentation (VOS) only consider single-shot videos. To address this challenge, we collected a new dataset-YouMVaS-of 200 popular YouTube videos spanning ten genres, where each video is on average five minutes long and with 75 shots. We selected recurring actors and annotated 431K segmentation masks at a frame rate of six, exceeding previous datasets in average video duration, object variation, and narrative structure complexity. We incorporated good practices of model architecture design, memory management, and multi-shot tracking into an existing video segmentation method to build competitive baseline methods. Through error analysis, we found that these baselines still fail to cope with cross-shot appearance variation on our YouMVOS dataset. Thus, our dataset poses new challenges in multi-shot segmentation towards better video analysis. Data, code, and pre-trained models are available at https://donglaiw.github.io/proj/youMVOS

Evaluation practices for image super-resolution (SR) use a single-value metric, the PSNR or SSIM, to determine model performance. This provides little insight into the source of errors and model behavior. Therefore, it is beneficial to move beyond the conventional approach and reconceptualize evaluation with interpretability as our main priority. We focus on a thorough error analysis from a variety of perspectives. Our key contribution is to leverage a texture classifier, which enables us to assign patches with semantic labels, to identify the source of SR errors both globally and locally. We then use this to determine (a) the semantic alignment of SR datasets, (b) how SR models perform on each label, (c) to what extent high-resolution (HR) and SR patches semantically correspond, and more. Through these different angles, we are able to highlight potential pitfalls and blindspots. Our overall investigation highlights numerous unexpected insights. We hope this work serves as an initial step for debugging blackbox SR networks.

Guided depth super-resolution (GDSR) is an essential topic in multi-modal image processing, which reconstructs high-resolution (HR) depth maps from low-resolution ones collected with suboptimal conditions with the help of HR RGB images of the same scene. To solve the challenges in interpreting the working mechanism, extracting cross-modal features and RGB texture over-transferred, we propose a novel Discrete Cosine Transform Network (DCTNet) to alleviate the problems from three aspects. First, the Discrete Cosine Transform (DCT) module reconstructs the multi-channel HR depth features by using DCT to solve the channel-wise optimization problem derived from the image domain. Second, we introduce a semi-coupled feature extraction module that uses shared convolutional kernels to extract common information and private kernels to extract modality-specific information. Third, we employ an edge attention mechanism to highlight the contours informative for guided upsampling. Extensive quantitative and qualitative evaluations demonstrate the effectiveness of our DCTNet, which outperforms previous state-of-the-art methods with a relatively small number of parameters. The code is available at https://github.com/Zhaozixiang1228/GDSR-DCTNet.

Abstract The field of connectomics aims to reconstruct the wiring diagram of Neurons and synapses to enable new insights into the workings of the brain. Reconstructing and analyzing the Neuronal connectivity, however, relies on many individual steps, starting from high‐resolution data acquisition to automated segmentation, proofreading, interactive data exploration, and circuit analysis. All of these steps have to handle large and complex datasets and rely on or benefit from integrated visualization methods. In this state‐of‐the‐art report, we describe visualization methods that can be applied throughout the connectomics pipeline, from data acquisition to circuit analysis. We first define the different steps of the pipeline and focus on how visualization is currently integrated into these steps. We also survey open science initiatives in connectomics, including usable open‐source tools and publicly available datasets. Finally, we discuss open challenges and possible future directions of this exciting research field.

Abstract High‐resolution electron microscopy imaging allows neuroscientists to reconstruct not just entire cells but individual cell substructures (i.e., cell organelles) as well. Based on these data, scientists hope to get a better understanding of brain function and development through detailed analysis of local organelle neighborhoods. In‐depth analyses require efficient and scalable comparison of a varying number of cell organelles, ranging from two to hundreds of local spatial neighborhoods. Scientists need to be able to analyze the 3D morphologies of organelles, their spatial distributions and distances, and their spatial correlations. We have designed Barrio as a configurable framework that scientists can adjust to their preferred workflow, visualizations, and supported user interactions for their specific tasks and domain questions. Furthermore, Barrio provides a scalable comparative visualization approach for spatial neighborhoods that automatically adjusts visualizations based on the number of structures to be compared. Barrio supports small multiples of spatial 3D views as well as abstract quantitative views, and arranges them in linked and juxtaposed views. To adapt to new domain‐specific analysis scenarios, we allow the definition of individualized visualizations and their parameters for each analysis session. We present an in‐depth case study for mitochondria analysis in neuronal tissue and demonstrate the usefulness of Barrio in a qualitative user study with neuroscientists.

The training data distribution is often biased towards objects in certain orientations and illumination conditions. While humans have a remarkable capability of recognizing objects in out-of-distribution (OoD) orientations and illuminations, Deep Neural Networks (DNNs) severely suffer in this case, even when large amounts of training examples are available. Neurons that are invariant to orientations and illuminations have been proposed as a neural mechanism that could facilitate OoD generalization, but it is unclear how to encourage the emergence of such invariant neurons. In this paper, we investigate three different approaches that lead to the emergence of invariant neurons and substantially improve DNNs in recognizing objects in OoD orientations and illuminations. Namely, these approaches are (i) training much longer after convergence of the in-distribution (InD) validation accuracy, i.e., late-stopping, (ii) tuning the momentum parameter of the batch normalization layers, and (iii) enforcing invariance of the neural activity in an intermediate layer to orientation and illumination conditions. Each of these approaches substantially improves the DNN's OoD accuracy (more than 20% in some cases). We report results in four datasets: two datasets are modified from the MNIST and iLab datasets, and the other two are novel (one of 3D rendered cars and another of objects taken from various controlled orientations and illumination conditions). These datasets allow to study the effects of different amounts of bias and are challenging as DNNs perform poorly in OoD conditions. Finally, we demonstrate that even though the three approaches focus on different aspects of DNNs, they all tend to lead to the same underlying neural mechanism to enable OoD accuracy gains - individual neurons in the intermediate layers become invariant to OoD orientations and illuminations. We anticipate this study to be a basis for further improvement of deep neural networks' OoD generalization performance, which is highly demanded to achieve safe and fair AI applications.

As connectomic datasets exceed hundreds of terabytes in size, accurate and efficient skeleton generation of the label volumes has evolved into a critical component of the computation pipeline used for analysis, evaluation, visualization, and error correction. We propose a novel topological thinning strategy that uses biological-constraints to produce accurate centerlines from segmented neuronal volumes while still maintaining biologically relevant properties. Current methods are either agnostic to the underlying biology, have non-linear running times as a function of the number of input voxels, or both. First, we eliminate from the input segmentation biologically-infeasible bubbles, pockets of voxels incorrectly labeled within a neuron, to improve segmentation accuracy, allow for more accurate centerlines, and increase processing speed. Next, a Convolutional Neural Network (CNN) detects cell bodies from the input segmentation, allowing us to anchor our skeletons to the somata. Lastly, a synapse-aware topological thinning approach produces expressive skeletons for each neuron with a nearly one-to-one correspondence between endpoints and synapses. We simultaneously estimate geometric properties of neurite width and geodesic distance between synapse and cell body, improving accuracy by 47.5% and 62.8% over baseline methods. We separate the skeletonization process into a series of computation steps, leveraging data-parallel strategies to increase throughput significantly. We demonstrate our results on over 1250 neurons and neuron fragments from three different species, processing over one million voxels per second per CPU with linear scalability.

Inspection of tissues using a light microscope is the primary method of diagnosing many diseases, notably cancer. Highly multiplexed tissue imaging builds on this foundation, enabling the collection of up to 60 channels of molecular information plus cell and tissue morphology using antibody staining. This provides unique insight into disease biology and promises to help with the design of patient-specific therapies. However, a substantial gap remains with respect to visualizing the resulting multivariate image data and effectively supporting pathology workflows in digital environments on screen. We, therefore, developed Scope2Screen, a scalable software system for focus+context exploration and annotation of whole-slide, high-plex, tissue images. Our approach scales to analyzing 100GB images of 109or more pixels per channel, containing millions of individual cells. A multidisciplinary team of visualization experts, microscopists, and pathologists identified key image exploration and annotation tasks involving finding, magnifying, quantifying, and organizing regions of interest (ROIs) in an intuitive and cohesive manner. Building on a scope-to-screen metaphor, we present interactive lensing techniques that operate at single-cell and tissue levels. Lenses are equipped with task-specific functionality and descriptive statistics, making it possible to analyze image features, cell types, and spatial arrangements (neighborhoods) across image channels and scales. A fast sliding-window search guides users to regions similar to those under the lens; these regions can be analyzed and considered either separately or as part of a larger image collection. A novel snapshot method enables linked lens configurations and image statistics to be saved, restored, and shared with these regions. We validate our designs with domain experts and apply Scope2Screen in two case studies involving lung and colorectal cancers to discover cancer-relevant image features.

Blazars are celestial bodies of high interest to astronomers. In particular, through the analysis of photometric and polarimetric observations of blazars, astronomers aim to understand the physics of the blazar's relativistic jet. However, it is challenging to recognize correlations and time variations of the observed polarization, intensity, and color of the emitted light. In our prior study, we proposed TimeTubes to visualize a blazar dataset as a 3D volumetric tube. In this paper, we build primarily on the TimeTubes representation of blazar datasets to present a new visual analytics environment named TimeTubesX, into which we have integrated sophisticated feature and pattern detection techniques for effective location of observable and recurring time variation patterns in long-term, multi-dimensional datasets. Automatic feature extraction detects time intervals corresponding to well-known blazar behaviors. Dynamic visual querying allows users to search long-term observations for time intervals similar to a time interval of interest (query-by-example) or a sketch of temporal patterns (query-by-sketch). Users are also allowed to build up another visual query guided by the time interval of interest found in the previous process and refine the results. We demonstrate how TimeTubesX has been used successfully by domain experts for the detailed analysis of blazar datasets and report on the results.

Table2Text systems generate textual output based on structured data utilizing machine learning. These systems are essential for fluent natural language interfaces in tools such as virtual assistants; however, left to generate freely these ML systems often produce misleading or unexpected outputs. GenNI (Generation Negotiation Interface) is an interactive visual system for high-level human-AI collaboration in producing descriptive text. The tool utilizes a deep learning model designed with explicit control states. These controls allow users to globally constrain model generations, without sacrificing the representation power of the deep learning models. The visual interface makes it possible for users to interact with AI systems following a Refine-Forecast paradigm to ensure that the generation system acts in a manner human users find suitable. We report multiple use cases on two experiments that improve over uncontrolled generation approaches, while at the same time providing fine-grained control. A demo and source code are available at https://genni.vizhub.ai.

The base learners and labeled samples (shots) in an ensemble few-shot classifier greatly affect the model performance. When the performance is not satisfactory, it is usually difficult to understand the underlying causes and make improvements. To tackle this issue, we propose a visual analysis method, FSLDiagnotor. Given a set of base learners and a collection of samples with a few shots, we consider two problems: 1) finding a subset of base learners that well predict the sample collections; and 2) replacing the low-quality shots with more representative ones to adequately represent the sample collections. We formulate both problems as sparse subset selection and develop two selection algorithms to recommend appropriate learners and shots, respectively. A matrix visualization and a scatterplot are combined to explain the recommended learners and shots in context and facilitate users in adjusting them. Based on the adjustment, the algorithm updates the recommendation results for another round of improvement. Two case studies are conducted to demonstrate that FSLDiagnotor helps build a few-shot classifier efficiently and increases the accuracy by 12% and 21%, respectively.

Widely used in news, business, and educational media, infographics are handcrafted to effectively communicate messages about complex and often abstract topics including `ways to conserve the environment' and `coronavirus prevention'. The computational understanding of infographics required for future applications like automatic captioning, summarization, search, and question-answering, will depend on being able to parse the visual and textual elements contained within. However, being composed of stylistically and semantically diverse visual and textual elements, infographics pose challenges for current A.I. systems. While automatic text extraction works reasonably well on infographics, standard object detection algorithms fail to identify the stand-alone visual elements in infographics that we refer to as `icons'. In this paper, we propose a novel approach to train an object detector using synthetically-generated data, and show that it succeeds at generalizing to detecting icons within in-the-wild infographics. We further pair our icon detection approach with an icon classifier and a state-of-the-art text detector to demonstrate three demo applications: topic prediction, multi-modal summarization, and multi-modal search. Parsing the visual and textual elements within infographics provides us with the first steps towards automatic infographic understanding.

Crowdsourced design feedback systems are emerging resources for getting large amounts of feedback in a short period of time. Traditionally, the feedback comes in the form of a declarative statement, which often contains positive or negative sentiment. Prior research has shown that overly negative or positive sentiment can strongly influence the perceived usefulness and acceptance of feedback and, subsequently, lead to ineffective design revisions. To enhance the effectiveness of crowdsourced design feedback, we investigate a new approach for mitigating the effects of negative or positive feedback by combining open-ended and thought-provoking questions with declarative feedback statements. We conducted two user studies to assess the effects of question-based feedback on the sentiment and quality of design revisions in the context of graphic design. We found that crowdsourced question-based feedback contains more neutral sentiment than statement-based feedback. Moreover, we provide evidence that presenting feedback as questions followed by statements leads to better design revisions than question- or statement-based feedback alone.

We present an observational study to compare co-located and situated real-time visualizations in basketball free-throw training. Our goal is to understand the advantages and concerns of applying immersive visualization to real-world skill-based sports training and to provide insights for designing AR sports training systems. We design both a situated 3D visualization on a head-mounted display and a 2D visualization on a co-located display to provide immediate visual feedback on a player’s shot performance. Using a within-subject study design with experienced basketball shooters, we characterize user goals, report on qualitative training experiences, and compare the quantitative training results. Our results show that real-time visual feedback helps athletes refine subsequent shots. Shooters in our study achieve greater angle consistency with our visual feedback. Furthermore, AR visualization promotes an increased focus on body form in athletes. Finally, we present suggestions for the design of future sports AR studies.

Quantifying user performance with metrics such as time and accuracy does not show the whole picture when researchers evaluate complex, interactive visualization tools. In such systems, performance is often influenced by different analysis strategies that statistical analysis methods cannot account for. To remedy this lack of nuance, we propose a novel analysis methodology for evaluating complex interactive visualizations at scale. We implement our analysis methods in reVISit, which enables analysts to explore participant interaction performance metrics and responses in the context of users’ analysis strategies. Replays of participant sessions can aid in identifying usability problems during pilot studies and make individual analysis processes salient. To demonstrate the applicability of reVISit to visualization studies, we analyze participant data from two published crowdsourced studies. Our findings show that reVISit can be used to reveal and describe novel interaction patterns, to analyze performance differences between different analysis strategies, and to validate or challenge design decisions.

The reflectance field of a face describes the reflectance properties responsible for complex lighting effects including diffuse, specular, inter-reflection and self shadowing. Most existing methods for estimating the face reflectance from a monocular image assume faces to be diffuse with very few approaches adding a specular component. This still leaves out important perceptual aspects of reflectance such as higher-order global illumination effects and self-shadowing. We present a new neural representation for face reflectance where we can estimate all components of the reflectance responsible for the final appearance from a monocular image. Instead of modeling each component of the reflectance separately using parametric models, our neural representation allows us to generate a basis set of faces in a geometric deformation-invariant space, parameterized by the input light direction, viewpoint and face geometry. We learn to reconstruct this reflectance field of a face just from a monocular image, which can be used to render the face from any viewpoint in any light condition. Our method is trained on a light-stage dataset, which captures 300 people illuminated with 150 light conditions from 8 viewpoints. We show that our method outperforms existing monocular reflectance reconstruction methods due to better capturing of physical effects, such as sub-surface scattering, specularities, self-shadows and other higher-order effects.

Deep convolutional neural networks (CNNs) have pushed forward the frontier of super-resolution (SR) research. However, current CNN models exhibit a major flaw: they are biased towards learning low-frequency signals. This bias becomes more problematic for the image SR task which targets reconstructing all fine details and image textures. To tackle this challenge, we propose to improve the learning of high-frequency features both locally and globally and introduce two novel architectural units to existing SR models. Specifically, we propose a dynamic highpass filtering (HPF) module that locally applies adaptive filter weights for each spatial location and channel group to preserve high-frequency signals. We also propose a matrix multi-spectral channel attention (MMCA) module that predicts the attention map of features decomposed in the frequency domain. This module operates in a global context to adaptively recalibrate feature responses at different frequencies. Extensive qualitative and quantitative results demonstrate that our proposed modules achieve better accuracy and visual improvements against state-of-the-art methods on several benchmark datasets.

Deep convolutional neural networks (CNNs) are achieving great successes for image super-resolution (SR), where global context is crucial for accurate restoration. However, the basic convolutional layer in CNNs is designed to extract local patterns, lacking the ability to model global context. With global context information, lots of efforts have been devoted to augmenting SR networks, especially by global feature interaction methods. These works incorporate the global context into local feature representation. However, recent advances in neuroscience show that it is necessary for the neurons to dynamically modulate their functions according to context, which is neglected in most CNN based SR methods. Motivated by those observations and analyses, we propose context reasoning attention network (CRAN) to modulate the convolution kernel according to the global context adaptively. Specifically, we extract global context descriptors, which are further enhanced with semantic reasoning. Channel and spatial interactions are then introduced to generate context reasoning attention mask, which is applied to modify the convolution kernel adaptively. Such a modulated convolution layer is utilized as basic component to build the blocks and networks. Extensive experiments on benchmark datasets with multiple degradation models show that CRAN obtains superior results and favorable trade-off between performance and model complexity.

Segmenting 3D cell nuclei from microscopy image volumes is critical for biological and clinical analysis, enabling the study of cellular expression patterns and cell lineages. However, current datasets for neuronal nuclei usually contain volumes smaller than $10^{\text{-}3}\ mm^3$ with fewer than 500 instances per volume, unable to reveal the complexity in large brain regions and restrict the investigation of neuronal structures. In this paper, we have pushed the task forward to the sub-cubic millimeter scale and curated the NucMM dataset with two fully annotated volumes: one $0.1\ mm^3$ electron microscopy (EM) volume containing nearly the entire zebrafish brain with around 170,000 nuclei; and one $0.25\ mm^3$ micro-CT (uCT) volume containing part of a mouse visual cortex with about 7,000 nuclei. With two imaging modalities and significantly increased volume size and instance numbers, we discover a great diversity of neuronal nuclei in appearance and density, introducing new challenges to the field. We also perform a statistical analysis to illustrate those challenges quantitatively. To tackle the challenges, we propose a novel hybrid-representation learning model that combines the merits of foreground mask, contour map, and signed distance transform to produce high-quality 3D masks. The benchmark comparisons on the NucMM dataset show that our proposed method significantly outperforms state-of-the-art nuclei segmentation approaches. Code and data are available at https://connectomics-bazaar.github.io/proj/nucMM/index.html.

Existing deep learning real denoising methods require a large amount of noisy-clean image pairs for supervision. Nonetheless, capturing a real noisy-clean dataset is an unacceptable expensive and cumbersome procedure. To alleviate this problem, this work investigates how to generate realistic noisy images. Firstly, we formulate a simple yet reasonable noise model that treats each real noisy pixel as a random variable. This model splits the noisy image generation problem into two sub-problems: image domain alignment and noise domain alignment. Subsequently, we propose a novel framework, namely Pixel-level Noise-aware Generative Adversarial Network (PNGAN). PNGAN employs a pre-trained real denoiser to map the fake and real noisy images into a nearly noise-free solution space to perform image domain alignment. Simultaneously, PNGAN establishes a pixel-level adversarial training to conduct noise domain alignment. Additionally, for better noise fitting, we present an efficient architecture Simple Multi-scale Network (SMNet) as the generator. Qualitative validation shows that noise generated by PNGAN is highly similar to real noise in terms of intensity and distribution. Quantitative experiments demonstrate that a series of denoisers trained with the generated noisy images achieve state-of-the-art (SOTA) results on four real denoising benchmarks.

Abstract A connectivity graph of neurons at the resolution of single synapses provides scientists with a tool for understanding the nervous system in health and disease. Recent advances in automatic image segmentation and synapse prediction in electron microscopy (EM) datasets of the brain have made reconstructions of neurons possible at the nanometer scale. However, automatic segmentation sometimes struggles to segment large neurons correctly, requiring human effort to proofread its output. General proofreading involves inspecting large volumes to correct segmentation errors at the pixel level, a visually intensive and time‐consuming process. This paper presents the design and implementation of an analytics framework that streamlines proofreading, focusing on connectivity‐related errors. We accomplish this with automated likely‐error detection and synapse clustering that drives the proofreading effort with highly interactive 3D visualizations. In particular, our strategy centers on proofreading the local circuit of a single cell to ensure a basic level of completeness. We demonstrate our framework's utility with a user study and report quantitative and subjective feedback from our users. Overall, users find the framework more efficient for proofreading, understanding evolving graphs, and sharing error correction strategies.

Abstract This paper surveys visualization and interaction techniques for geospatial networks from a total of 95 papers. Geospatial networks are graphs where nodes and links can be associated with geographic locations. Examples can include social networks, trade and migration, as well as traffic and transport networks. Visualizing geospatial networks poses numerous challenges around the integration of both network and geographical information as well as additional information such as node and link attributes, time and uncertainty. Our overview analyses existing techniques along four dimensions: (i) the representation of geographical information, (ii) the representation of network information, (iii) the visual integration of both and (iv) the use of interaction. These four dimensions allow us to discuss techniques with respect to the trade‐offs they make between showing information across all these dimensions and how they solve the problem of showing as much information as necessary while maintaining readability of the visualization. https://geonetworks.github.io .

Photorealistic editing of head portraits is a challenging task as humans are very sensitive to inconsistencies in faces. We present an approach for high-quality intuitive editing of the camera viewpoint and scene illumination (parameterised with an environment map) in a portrait image. This requires our method to capture and control the full reflectance field of the person in the image. Most editing approaches rely on supervised learning using training data captured with setups such as light and camera stages. Such datasets are expensive to acquire, not readily available and do not capture all the rich variations of in-the-wild portrait images. In addition, most supervised approaches only focus on relighting, and do not allow camera viewpoint editing. Thus, they only capture and control a subset of the reflectance field. Recently, portrait editing has been demonstrated by operating in the generative model space of StyleGAN. While such approaches do not require direct supervision, there is a significant loss of quality when compared to the supervised approaches. In this paper, we present a method which learns from limited supervised training data. The training images only include people in a fixed neutral expression with eyes closed, without much hair or background variations. Each person is captured under 150 one-light-at-a-time conditions and under 8 camera poses. Instead of training directly in the image space, we design a supervised problem which learns transformations in the latent space of StyleGAN. This combines the best of supervised learning and generative adversarial modeling. We show that the StyleGAN prior allows for generalisation to different expressions, hairstyles and backgrounds. This produces high-quality photorealistic results for in-the-wild images and significantly outperforms existing methods. Our approach can edit the illumination and pose simultaneously, and runs at interactive rates.

When adding a photo onto a graphic design, professional graphic designers often adjust its colors based on some target colors obtained from the brand or product to make the entire design more memorable to audiences and establish a consistent brand identity. However, adjusting the colors of a photo in the context of a graphic design is a difficult task, with two major challenges: (1) Locality: The color is often adjusted locally to preserve the semantics and atmosphere of the original image; and (2) Naturalness: The modified region needs to be carefully chosen and recolored to obtain a semantically valid and visually natural result. To address these challenges, we propose a learning-based approach to photo color adjustment for graphic designs, which maps an input photo along with the target colors to a recolored result. Our method decomposes the color adjustment process into two successive stages: modifiable region selection and target color propagation. The first stage aims to solve the core, challenging problem of which local image region(s) should be adjusted, which requires not only a common sense of colors appearing in our visual world but also understanding of subtle visual design heuristics. To this end, we capitalize on both natural photos and graphic designs to train a region selection network, which detects the most likely regions to be adjusted to the target colors. The second stage trains a recoloring network to naturally propagate the target colors in the detected regions. Through extensive experiments and a user study, we demonstrate the effectiveness of our selective region-based photo recoloring framework.

Small multiples are miniature representations of visual information used generically across many domains. Handling large numbers of small multiples imposes challenges on many analytic tasks like inspection, comparison, navigation, or annotation. To address these challenges, we developed a framework and implemented a library called PILlNG.JS for designing interactive piling interfaces. Based on the piling metaphor, such interfaces afford flexible organization, exploration, and comparison of large numbers of small multiples by interactively aggregating visual objects into piles. Based on a systematic analysis of previous work, we present a structured design space to guide the design of visual piling interfaces. To enable designers to efficiently build their own visual piling interfaces, PILlNG.JS provides a declarative interface to avoid having to write low-level code and implements common aspects of the design space. An accompanying GUI additionally supports the dynamic configuration of the piling interface. We demonstrate the expressiveness of PILlNG.JS with examples from machine learning, immunofluorescence microscopy, genomics, and public health.

Computing and visualizing features in fluid flow often depends on the observer, or reference frame, relative to which the input velocity field is given. A desired property of feature detectors is therefore that they are objective, meaning independent of the input reference frame. However, the standard definition of objectivity is only given for Euclidean domains and cannot be applied in curved spaces. We build on methods from mathematical physics and Riemannian geometry to generalize objectivity to curved spaces, using the powerful notion of symmetry groups as the basis for definition. From this, we develop a general mathematical framework for the objective computation of observer fields for curved spaces, relative to which other computed measures become objective. An important property of our framework is that it works intrinsically in 2D, instead of in the 3D ambient space. This enables a direct generalization of the 2D computation via optimization of observer fields in flat space to curved domains, without having to perform optimization in 3D. We specifically develop the case of unsteady 2D geophysical flows given on spheres, such as the Earth. Our observer fields in curved spaces then enable objective feature computation as well as the visualization of the time evolution of scalar and vector fields, such that the automatically computed reference frames follow moving structures like vortices in a way that makes them appear to be steady.

data has no natural scale and so interactive data visualizations must provide techniques to allow the user to choose their viewpoint and scale. Such techniques are well established in desktop visualization tools. The two most common techniques are zoom+pan and overview+detail. However, how best to enable the analyst to navigate and view abstract data at different levels of scale in immersive environments has not previously been studied. We report the findings of the first systematic study of immersive navigation techniques for 3D scatterplots. We tested four conditions that represent our best attempt to adapt standard 2D navigation techniques to data visualization in an immersive environment while still providing standard immersive navigation techniques through physical movement and teleportation. We compared room-sized visualization versus a zooming interface, each with and without an overview. We find significant differences in participants' response times and accuracy for a number of standard visual analysis tasks. Both zoom and overview provide benefits over standard locomotion support alone (i.e., physical movement and pointer teleportation). However, which variation is superior, depends on the task. We obtain a more nuanced understanding of the results by analyzing them in terms of a time-cost model for the different components of navigation: way-finding, travel, number of travel steps, and context switching.

Infographics are engaging visual representations that tell an informative story using a fusion of data and graphical elements. The large variety of infographic design poses a challenge for their high-level analysis. We use the concept of Visual Information Flow (VIF), which is the underlying semantic structure that links graphical elements to convey the information and story to the user. To explore VIF, we collected a repository of over 13K infographics. We use a deep neural network to identify visual elements related to information, agnostic to their various artistic appearances. We construct the VIF by automatically chaining these visual elements together based on Gestalt principles. Using this analysis, we characterize the VIF design space by a taxonomy of 12 different design patterns. Exploring in a real-world infographic dataset, we discuss the design space and potentials of VIF in light of this taxonomy.

Compound icons are prevalent on signs, webpages, and infographics, effectively conveying complex and abstract concepts, such as "no smoking" and "health insurance", with simple graphical representations. However, designing such icons requires experience and creativity, in order to efficiently navigate the semantics, space, and style features of icons. In this paper, we aim to automate the process of generating icons given compound concepts, to facilitate rapid compound icon creation and ideation. Informed by ethnographic interviews with professional icon designers, we have developed ICONATE, a novel system that automatically generates compound icons based on textual queries and allows users to explore and customize the generated icons. At the core of ICONATE is a computational pipeline that automatically finds commonly used icons for sub-concepts and arranges them according to inferred conventions. To enable the pipeline, we collected a new dataset, Compicon1k, consisting of 1000 compound icons annotated with semantic labels (i.e., concepts). Through user studies, we have demonstrated that our tool is able to automate or accelerate the compound icon design process for both novices and professionals.

We present Peax, a novel feature-based technique for interactive visual pattern search in sequential data, like time series or data mapped to a genome sequence. Visually searching for patterns by similarity is often challenging because of the large search space, the visual complexity of patterns, and the user's perception of similarity. For example, in genomics, researchers try to link patterns in multivariate sequential data to cellular or pathogenic processes, but a lack of ground truth and high variance makes automatic pattern detection unreliable. We have developed a convolutional autoencoder for unsupervised representation learning of regions in sequential data that can capture more visual details of complex patterns compared to existing similarity measures. Using this learned representation as features of the sequential data, our accompanying visual query system enables interactive feedback-driven adjustments of the pattern search to adapt to the users' perceived similarity. Using an active learning sampling strategy, Peax collects user-generated binary relevance feedback. This feedback is used to train a model for binary classification, to ultimately find other regions that exhibit patterns similar to the search target. We demonstrate Peax's features through a case study in genomics and report on a user study with eight domain experts to assess the usability and usefulness of Peax. Moreover, we evaluate the effectiveness of the learned feature representation for visual similarity search in two additional user studies. We find that our models retrieve significantly more similar patterns than other commonly used techniques.

Matrix representations are one of the main established and empirically proven to be effective visualization techniques for relational (or network) data. However, matrices-similar to node-link diagrams-are most effective if their layout reveals the underlying data topology. Given the many developed algorithms, a practical problem arises: "Which matrix reordering algorithm should I choose for my dataset at hand?" To make matters worse, different reordering algorithms applied to the same dataset may let significantly different visual matrix patterns emerge. This leads to the question of trustworthiness and explainability of these fully automated, often heuristic, black-box processes. We present GUIRO, a Visual Analytics system that helps novices, network analysts, and algorithm designers to open the black-box. Users can investigate the usefulness and expressiveness of 70 accessible matrix reordering algorithms. For network analysts, we introduce a novel model space representation and two interaction techniques for a user-guided reordering of rows or columns, and especially groups thereof (submatrix reordering). These novel techniques contribute to the understanding of the global and local dataset topology. We support algorithm designers by giving them access to 16 reordering quality metrics and visual exploration means for comparing reordering implementations on a row/column permutation level. We evaluated GUIRO in a guided explorative user study with 12 subjects, a case study demonstrating its usefulness in a real-world scenario, and through an expert study gathering feedback on our design decisions. We found that our proposed methods help even inexperienced users to understand matrix patterns and allow a user-guided steering of reordering algorithms. GUIRO helps to increase the transparency of matrix reordering algorithms, thus helping a broad range of users to get a better insight into the complex reordering process, in turn supporting data and reordering algorithm insights.

Automation of tasks can have critical consequences when humans lose agency over decision processes. Deep learning models are particularly susceptible since current black-box approaches lack explainable reasoning. We argue that both the visual interface and model structure of deep learning systems need to take into account interaction design. We propose a framework of collaborative semantic inference (CSI) for the co-design of interactions and models to enable visual collaboration between humans and algorithms. The approach exposes the intermediate reasoning process of models which allows semantic interactions with the visual metaphors of a problem, which means that a user can both understand and control parts of the model reasoning process. We demonstrate the feasibility of CSI with a co-designed case study of a document summarization system.

Facetto is a scalable visual analytics application that is used to discover single-cell phenotypes in high-dimensional multi-channel microscopy images of human tumors and tissues. Such images represent the cutting edge of digital histology and promise to revolutionize how diseases such as cancer are studied, diagnosed, and treated. Highly multiplexed tissue images are complex, comprising 109 or more pixels, 60-plus channels, and millions of individual cells. This makes manual analysis challenging and error-prone. Existing automated approaches are also inadequate, in large part, because they are unable to effectively exploit the deep knowledge of human tissue biology available to anatomic pathologists. To overcome these challenges, Facetto enables a semi-automated analysis of cell types and states. It integrates unsupervised and supervised learning into the image and feature exploration process and offers tools for analytical provenance. Experts can cluster the data to discover new types of cancer and immune cells and use clustering results to train a convolutional neural network that classifies new cells accordingly. Likewise, the output of classifiers can be clustered to discover aggregate patterns and phenotype subsets. We also introduce a new hierarchical approach to keep track of analysis steps and data subsets created by users; this assists in the identification of cell types. Users can build phenotype trees and interact with the resulting hierarchical structures of both high-dimensional feature and image spaces. We report on use-cases in which domain scientists explore various large-scale fluorescence imaging datasets. We demonstrate how Facetto assists users in steering the clustering and classification process, inspecting analysis results, and gaining new scientific insights into cancer biology.

We present Scalable Insets, a technique for interactively exploring and navigating large numbers of annotated patterns in multiscale visualizations such as gigapixel images, matrices, or maps. Exploration of many but sparsely-distributed patterns in multiscale visualizations is challenging as visual representations change across zoom levels, context and navigational cues get lost upon zooming, and navigation is time consuming. Our technique visualizes annotated patterns too small to be identifiable at certain zoom levels using insets, i.e., magnified thumbnail views of the annotated patterns. Insets support users in searching, comparing, and contextualizing patterns while reducing the amount of navigation needed. They are dynamically placed either within the viewport or along the boundary of the viewport to offer a compromise between locality and context preservation. Annotated patterns are interactively clustered by location and type. They are visually represented as an aggregated inset to provide scalable exploration within a single viewport. In a controlled user study with 18 participants, we found that Scalable Insets can speed up visual search and improve the accuracy of pattern comparison at the cost of slower frequency estimation compared to a baseline technique. A second study with 6 experts in the field of genomics showed that Scalable Insets is easy to learn and provides first insights into how Scalable Insets can be applied in an open-ended data exploration scenario.

Many personal informatics systems allow people to collect and manage personal data and reflect more deeply about themselves. However, these tools rarely offer ways to customize how the data is visualized. In this work, we investigate the question of how to enable people to determine the representation of their data. We analyzed the Dear Data project to gain insights into the design elements of personal visualizations. We developed DataSelfie, a novel system that allows individuals to gather personal data and design custom visuals to represent the collected data. We conducted a user study to evaluate the usability of the system as well as its potential for individual and collaborative sensemaking of the data.

Comics are an entertaining and familiar medium for presenting compelling stories about data. However, existing visualization authoring tools do not leverage this expressive medium. In this paper, we seek to incorporate elements of comics into the construction of data-driven stories about dynamic networks. We contribute DataToon, a flexible data comic storyboarding tool that blends analysis and presentation with pen and touch interactions. A storyteller can use DataToon rapidly generate visualization panels, annotate them, and position them within a canvas to produce a visually compelling narrative. In a user study, participants quickly learned to use DataToon for producing data comics.

Most current state-of-the-art connectome reconstruction pipelines have two major steps: initial pixel-based segmentation with affinity prediction and watershed transform, and refined segmentation by merging over-segmented regions. These methods rely only on local context and are typically agnostic to the underlying biology. Since a few merge errors can lead to several incorrectly merged neuronal processes, these algorithms are currently tuned towards over-segmentation producing an overburden of costly proofreading. We propose a third step for connectomics reconstruction pipelines to refine an over-segmentation using both local and global context with an emphasis on adhering to the underlying biology. We first extract a graph from an input segmentation where nodes correspond to segment labels and edges indicate potential split errors in the over-segmentation. In order to increase throughput and allow for large-scale reconstruction, we employ biologically inspired geometric constraints based on neuron morphology to reduce the number of nodes and edges. Next, two neural networks learn these neuronal shapes to further aid the graph construction process. Lastly, we reformulate the region merging problem as a graph partitioning one to leverage global context. We demonstrate the performance of our approach on four real-world connectomics datasets with an average variation of information improvement of 21.3%.

Abstract The analysis of behavioral city dynamics, such as temporal patterns of visited places and citizens' mobility routines, is an essential task for urban and transportation planning. Social media applications such as Foursquare and Twitter provide access to large‐scale and up‐to‐date dynamic movement data that not only help to understand the social life and pulse of a city but also to maintain and improve urban infrastructure. However, the fast growth rate of this data poses challenges for conventional methods to provide up‐to‐date, flexible analysis. Therefore, planning authorities barely consider it. We present a system and design study to leverage social media data that assist urban and transportation planners to achieve better monitoring and analysis of city dynamics such as visited places and mobility patterns in large metropolitan areas. We conducted a goal‐and‐task analysis with urban planning experts. To address these goals, we designed a system with a scalable data monitoring back‐end and an interactive visual analytics interface. The monitoring component uses intelligent pre‐aggregation to allow dynamic queries in near real‐time. The visual analytics interface leverages unsupervised learning to reveal clusters, routines, and unusual behavior in massive data, allowing to understand patterns in time and space. We evaluated our approach based on a qualitative user study with urban planning experts which demonstrates that intuitive integration of advanced analytical tools with visual interfaces is pivotal in making behavioral city dynamics accessible to practitioners. Our interviews also revealed areas for future research.

Five years after the first state-of-the-art report on Commercial Visual Analytics Systems we present a reevaluation of the Big Data Analytics field. We build on the success of the 2012 survey, which was influential even beyond the boundaries of the InfoVis and Visual Analytics (VA) community. While the field has matured significantly since the original survey, we find that innovation and research-driven development are increasingly sacrificed to satisfy a wide range of user groups. We evaluate new product versions on established evaluation criteria, such as available features, performance, and usability, to extend on and assure comparability with the previous survey. We also investigate previously unavailable products to paint a more complete picture of the commercial VA landscape. Furthermore, we introduce novel measures, like suitability for specific user groups and the ability to handle complex data types, and undertake a new case study to highlight innovative features. We explore the achievements in the commercial sector in addressing VA challenges and propose novel developments that should be on systems' roadmaps in the coming years.

With the rapid increase in raw volume data sizes, such as terabyte-sized microscopy volumes, the corresponding segmentation label volumes have become extremely large as well. We focus on integer label data, whose efficient representation in memory, as well as fast random data access, pose an even greater challenge than the raw image data. Often, it is crucial to be able to rapidly identify which segments are located where, whether for empty space skipping for fast rendering, or for spatial proximity queries. We refer to this process as culling. In order to enable efficient culling of millions of labeled segments, we present a novel hybrid approach that combines deterministic and probabilistic representations of label data in a data-adaptive hierarchical data structure that we call the label list tree. In each node, we adaptively encode label data using either a probabilistic constant-time access representation for fast conservative culling, or a deterministic logarithmic-time access representation for exact queries. We choose the best data structures for representing the labels of each spatial region while building the label list tree. At run time, we further employ a novel query-adaptive culling strategy. While filtering a query down the tree, we prune it successively, and in each node adaptively select the representation that is best suited for evaluating the pruned query, depending on its size. We show an analysis of the efficiency of our approach with several large data sets from connectomics, including a brain scan with more than 13 million labeled segments, and compare our method to conventional culling approaches. Our approach achieves significant reductions in storage size as well as faster query times.

Convolutional neural networks can successfully perform many computer vision tasks on images. For visualization, how do CNNs perform when applied to graphical perception tasks? We investigate this question by reproducing Cleveland and McGill's seminal 1984 experiments, which measured human perception efficiency of different visual encodings and defined elementary perceptual tasks for visualization. We measure the graphical perceptual capabilities of four network architectures on five different visualization tasks and compare to existing and new human performance baselines. While under limited circumstances CNNs are able to meet or outperform human task performance, we find that CNNs are not currently a good model for human graphical perception. We present the results of these experiments to foster the understanding of how CNNs succeed and fail when applied to data visualizations.

This paper presents DXR, a toolkit for building immersive data visualizations based on the Unity development platform. Over the past years, immersive data visualizations in augmented and virtual reality (AR, VR) have been emerging as a promising medium for data sense-making beyond the desktop. However, creating immersive visualizations remains challenging, and often require complex low-level programming and tedious manual encoding of data attributes to geometric and visual properties. These can hinder the iterative idea-to-prototype process, especially for developers without experience in 3D graphics, AR, and VR programming. With DXR, developers can efficiently specify visualization designs using a concise declarative visualization grammar inspired by Vega-Lite. DXR further provides a GUI for easy and quick edits and previews of visualization designs in-situ, i.e., while immersed in the virtual world. DXR also provides reusable templates and customizable graphical marks, enabling unique and engaging visualizations. We demonstrate the flexibility of DXR through several examples spanning a wide range of applications.

Neural sequence-to-sequence models have proven to be accurate and robust for many sequence prediction tasks, and have become the standard approach for automatic translation of text. The models work with a five-stage blackbox pipeline that begins with encoding a source sequence to a vector space and then decoding out to a new target sequence. This process is now standard, but like many deep learning methods remains quite difficult to understand or debug. In this work, we present a visual analysis tool that allows interaction and "what if"-style exploration of trained sequence-to-sequence models through each stage of the translation process. The aim is to identify which patterns have been learned, to detect model errors, and to probe the model with counterfactual scenario. We demonstrate the utility of our tool through several real-world sequence-to-sequence use cases on large-scale models.

Automatic cell image segmentation methods in connectomics produce merge and split errors, which require correction through proofreading. Previous research has identified the visual search for these errors as the bottleneck in interactive proofreading. To aid error correction, we develop two classifiers that automatically recommend candidate merges and splits to the user. These classifiers use a convolutional neural network (CNN) that has been trained with errors in automatic segmentations against expert-labeled ground truth. Our classifiers detect potentially-erroneous regions by considering a large context region around a segmentation boundary. Corrections can then be performed by a user with yes/no decisions, which reduces variation of information 7.5× faster than previous proofreading methods. We also present a fully-automatic mode that uses a probability threshold to make merge/split decisions. Extensive experiments using the automatic approach and comparing performance of novice and expert users demonstrate that our method performs favorably against state-of-the-art proofreading methods on different connectomics datasets.

Motivation: Protein evolution spans time scales and its effects span the length of an organism. A web app named ProteomeVis is developed to provide a comprehensive view of protein evolution in the Saccharomyces cerevisiae and Escherichia coli proteomes. ProteomeVis interactively creates protein chain graphs, where edges between nodes represent structure and sequence similarities within user-defined ranges, to study the long time scale effects of protein structure evolution. The short time scale effects of protein sequence evolution are studied by sequence evolutionary rate (ER) correlation analyses with protein properties that span from the molecular to the organismal level. Results: We demonstrate the utility and versatility of ProteomeVis by investigating the distribution of edges per node in organismal protein chain universe graphs (oPCUGs) and putative ER determinants. S.cerevisiae and E.coli oPCUGs are scale-free with scaling constants of 1.79 and 1.56, respectively. Both scaling constants can be explained by a previously reported theoretical model describing protein structure evolution. Protein abundance most strongly correlates with ER among properties in ProteomeVis, with Spearman correlations of -0.49 (P-value < 10-10) and -0.46 (P-value < 10-10) for S.cerevisiae and E.coli, respectively. This result is consistent with previous reports that found protein expression to be the most important ER determinant. Availability and implementation: ProteomeVis is freely accessible at http://proteomevis.chem.harvard.edu. Supplementary information: Supplementary data are available at Bioinformatics online.

Bioinformatics (2018) https://doi.org/10.1093/bioinformatics/bty370 The publisher wishes to apologize for a mistake in Figure 1 in the above manuscript. The figure appears correctly below: The ProteomeVis web app after exploration has occurred. Hovering over asterisks located in the top-right corner of each panel opens an informational window. Pressing the help button (question mark symbol) located to the right of the Control Strip (A) leads to a quick tour. Alphabetical labels (A–E) are added to each component of ProteomeVis in this figure to introduce the tool in the text The paper has been corrected online.

Abstract The visualization community has developed to date many intuitions and understandings of how to judge thequalityof views in visualizing data. The computation of a visualization's quality and usefulness ranges from measuring clutter and overlap, up to the existence and perception of specific (visual) patterns. This survey attempts to report, categorize and unify the diverse understandings and aims to establish a common vocabulary that will enable a wide audience to understand their differences and subtleties. For this purpose, we present a commonly applicable quality metric formalization that should detail and relate all constituting parts of a quality metric. We organize our corpus of reviewed research papers along the data types established in the information visualization community: multi‐ and high‐dimensional, relational, sequential, geospatial and text data. For each data type, we select the visualization subdomains in which quality metrics are an active research field and report their findings, reason on the underlying concepts, describe goals and outline the constraints and requirements. One central goal of this survey is to provide guidance on future research opportunities for the field and outline how different visualization communities could benefit from each other by applying or transferring knowledge to their respective subdomain. Additionally, we aim to motivate the visualization community to compare computed measures to the perception of humans.

We present an effective blind image deblurring algorithm based on the dark channel prior. The motivation of this work is an interesting observation that the dark channel of blurred images is less sparse. While most patches in a clean image contain some dark pixels, this is not the case when they are averaged with neighboring ones by motion blur. This change in sparsity of the dark channel pixels is an inherent property of the motion blur process, which we prove mathematically and validate using image data. Enforcing sparsity of the dark channel thus helps blind deblurring in various scenarios such as natural, face, text, and low-illumination images. However, imposing sparsity of the dark channel introduces a non-convex non-linear optimization problem. In this work, we introduce a linear approximation to address this issue. Extensive experiments demonstrate that the proposed deblurring algorithm achieves the state-of-the-art results on natural images and performs favorably against methods designed for specific scenarios. In addition, we show that the proposed method can be applied to image dehazing.

We report on a controlled user study comparing three visualization environments for common 3D exploration. Our environments differ in how they exploit natural human perception and interaction capabilities. We compare an augmented-reality head-mounted display (Microsoft HoloLens), a handheld tablet, and a desktop setup. The novel head-mounted HoloLens display projects stereoscopic images of virtual content into a user's real world and allows for interaction in-situ at the spatial position of the 3D hologram. The tablet is able to interact with 3D content through touch, spatial positioning, and tangible markers, however, 3D content is still presented on a 2D surface. Our hypothesis is that visualization environments that match human perceptual and interaction capabilities better to the task at hand improve understanding of 3D visualizations. To better understand the space of display and interaction modalities in visualization environments, we first propose a classification based on three dimensions: perception, interaction, and the spatial and cognitive proximity of the two. Each technique in our study is located at a different position along these three dimensions. We asked 15 participants to perform four tasks, each task having different levels of difficulty for both spatial perception and degrees of freedom for interaction. Our results show that each of the tested environments is more effective for certain tasks, but that generally the desktop environment is still fastest and most precise in almost all cases.

Recent advances in data acquisition produce volume data of very high resolution and large size, such as terabyte-sized microscopy volumes. These data often contain many fine and intricate structures, which pose huge challenges for volume rendering, and make it particularly important to efficiently skip empty space. This paper addresses two major challenges: (1) The complexity of large volumes containing fine structures often leads to highly fragmented space subdivisions that make empty regions hard to skip efficiently. (2) The classification of space into empty and non-empty regions changes frequently, because the user or the evaluation of an interactive query activate a different set of objects, which makes it unfeasible to pre-compute a well-adapted space subdivision. We describe the novel SparseLeap method for efficient empty space skipping in very large volumes, even around fine structures. The main performance characteristic of SparseLeap is that it moves the major cost of empty space skipping out of the ray-casting stage. We achieve this via a hybrid strategy that balances the computational load between determining empty ray segments in a rasterization (object-order) stage, and sampling non-empty volume data in the ray-casting (image-order) stage. Before ray-casting, we exploit the fast hardware rasterization of GPUs to create a ray segment list for each pixel, which identifies non-empty regions along the ray. The ray-casting stage then leaps over empty space without hierarchy traversal. Ray segment lists are created by rasterizing a set of fine-grained, view-independent bounding boxes. Frame coherence is exploited by re-using the same bounding boxes unless the set of active objects changes. We show that SparseLeap scales better to large, sparse data than standard octree empty space skipping.

In this paper, we present story curves, a visualization technique for exploring and communicating nonlinear narratives in movies. A nonlinear narrative is a storytelling device that portrays events of a story out of chronological order, e.g., in reverse order or going back and forth between past and future events. Many acclaimed movies employ unique narrative patterns which in turn have inspired other movies and contributed to the broader analysis of narrative patterns in movies. However, understanding and communicating nonlinear narratives is a difficult task due to complex temporal disruptions in the order of events as well as no explicit records specifying the actual temporal order of the underlying story. Story curves visualize the nonlinear narrative of a movie by showing the order in which events are told in the movie and comparing them to their actual chronological order, resulting in possibly meandering visual patterns in the curve. We also present Story Explorer, an interactive tool that visualizes a story curve together with complementary information such as characters and settings. Story Explorer further provides a script curation interface that allows users to specify the chronological order of events in movies. We used Story Explorer to analyze 10 popular nonlinear movies and describe the spectrum of narrative patterns that we discovered, including some novel patterns not previously described in the literature. Feedback from experts highlights potential use cases in screenplay writing and analysis, education and film production. A controlled user study shows that users with no expertise are able to understand visual patterns of nonlinear narratives using story curves.

This paper presents an interactive visualization interface-HiPiler-for the exploration and visualization of regions-of-interest in large genome interaction matrices. Genome interaction matrices approximate the physical distance of pairs of regions on the genome to each other and can contain up to 3 million rows and columns with many sparse regions. Regions of interest (ROIs) can be defined, e.g., by sets of adjacent rows and columns, or by specific visual patterns in the matrix. However, traditional matrix aggregation or pan-and-zoom interfaces fail in supporting search, inspection, and comparison of ROIs in such large matrices. In HiPiler, ROIs are first-class objects, represented as thumbnail-like "snippets". Snippets can be interactively explored and grouped or laid out automatically in scatterplots, or through dimension reduction methods. Snippets are linked to the entire navigable genome interaction matrix through brushing and linking. The design of HiPiler is based on a series of semi-structured interviews with 10 domain experts involved in the analysis and interpretation of genome interaction matrices. We describe six exploration tasks that are crucial for analysis of interaction matrices and demonstrate how HiPiler supports these tasks. We report on a user study with a series of data exploration sessions with domain experts to assess the usability of HiPiler as well as to demonstrate respective findings in the data.

This paper presents Abstractocyte, a system for the visual analysis of astrocytes and their relation to neurons, in nanoscale volumes of brain tissue. Astrocytes are glial cells, i.e., non-neuronal cells that support neurons and the nervous system. The study of astrocytes has immense potential for understanding brain function. However, their complex and widely-branching structure requires high-resolution electron microscopy imaging and makes visualization and analysis challenging. Furthermore, the structure and function of astrocytes is very different from neurons, and therefore requires the development of new visualization and analysis tools. With Abstractocyte, biologists can explore the morphology of astrocytes using various visual abstraction levels, while simultaneously analyzing neighboring neurons and their connectivity. We define a novel, conceptual 2D abstraction space for jointly visualizing astrocytes and neurons. Neuroscientists can choose a specific joint visualization as a point in this space. Interactively moving this point allows them to smoothly transition between different abstraction levels in an intuitive manner. In contrast to simply switching between different visualizations, this preserves the visual context and correlations throughout the transition. Users can smoothly navigate from concrete, highly-detailed 3D views to simplified and abstracted 2D views. In addition to investigating astrocytes, neurons, and their relationships, we enable the interactive analysis of the distribution of glycogen, which is of high importance to neuroscientists. We describe the design of Abstractocyte, and present three case studies in which neuroscientists have successfully used our system to assess astrocytic coverage of synapses, glycogen distribution in relation to synapses, and astrocytic-mitochondria coverage.

Recurrent neural networks, and in particular long short-term memory (LSTM) networks, are a remarkably effective tool for sequence modeling that learn a dense black-box hidden representation of their sequential input. Researchers interested in better understanding these models have studied the changes in hidden state representations over time and noticed some interpretable patterns but also significant noise. In this work, we present LSTMVis, a visual analysis tool for recurrent neural networks with a focus on understanding these hidden state dynamics. The tool allows users to select a hypothesis input range to focus on local state changes, to match these states changes to similar patterns in a large data set, and to align these results with structural annotations from their domain. We show several use cases of the tool for analyzing specific hidden state properties on dataset containing nesting, phrase structure, and chord progressions, and demonstrate how the tool can be used to isolate patterns for further statistical analysis. We characterize the domain, the different stakeholders, and their goals and tasks. Long-term usage data after putting the tool online revealed great interest in the machine learning community.

Building robust classifiers trained on data susceptible to group or subject-specific variations is a challenging pattern recognition problem. We develop hierarchical Bayesian neural networks to capture subject-specific variations and share statistical strength across subjects. Leveraging recent work on learning Bayesian neural networks, we build fast, scalable algorithms for inferring the posterior distribution over all network weights in the hierarchy. We also develop methods for adapting our model to new subjects when a small number of subject-specific personalization data is available. Finally, we investigate active learning algorithms for interactively labeling personalization data in resource-constrained scenarios. Focusing on the problem of gesture recognition where inter-subject variations are commonplace, we demonstrate the effectiveness of our proposed techniques. We test our framework on three widely used gesture recognition datasets, achieving personalization performance competitive with the state-of-the-art.

Can overtrust in robots compromise physical security? We conducted a series of experiments in which a robot positioned outside a secure-access student dormitory asked passersby to assist it to gain access. We found individual participants were as likely to assist the robot in exiting the dormitory (40% assistance rate, 4/10 individuals) as in entering (19%, 3/16 individuals). Groups of people were more likely than individuals to assist the robot in entering (71%, 10/14 groups). When the robot was disguised as a food delivery agent for the fictional start-up Robot Grub, individuals were more likely to assist the robot in entering (76%, 16/21 individuals). Lastly, we found participants who identified the robot as a bomb threat demonstrated a trend toward assisting the robot (87%, 7/8 individuals, 6/7 groups). Thus, we demonstrate that overtrust---the unfounded belief that the robot does not intend to deceive or carry risk---can represent a significant threat to physical security at a university dormitory.

We present an algorithm to directly restore a clear highresolution image from a blurry low-resolution input. This problem is highly ill-posed and the basic assumptions for existing super-resolution methods (requiring clear input) and deblurring methods (requiring high-resolution input) no longer hold. We focus on face and text images and adopt a generative adversarial network (GAN) to learn a category-specific prior to solve this problem. However, the basic GAN formulation does not generate realistic high-resolution images. In this work, we introduce novel training losses that help recover fine details. We also present a multi-class GAN that can process multi-class image restoration tasks, i.e., face and text images, using a single generator network. Extensive experiments demonstrate that our method performs favorably against the state-of-the-art methods on both synthetic and real-world images at a lower computational cost.

Knowing where people look and click on visual designs can provide clues about how the designs are perceived, and where the most important or relevant content lies. The most important content of a visual design can be used for effective summarization or to facilitate retrieval from a database. We present automated models that predict the relative importance of different elements in data visualizations and graphic designs. Our models are neural networks trained on human clicks and importance annotations on hundreds of designs. We collected a new dataset of crowdsourced importance, and analyzed the predictions of our models with respect to ground truth importance and human eye movements. We demonstrate how such predictions of importance can be used for automatic design retargeting and thumbnailing. User studies with hundreds of MTurk participants validate that, with limited post-processing, our importance-driven applications are on par with, or outperform, current state-of-the-art methods, including natural image saliency. We also provide a demonstration of how our importance predictions can be built into interactive design tools to offer immediate feedback during the design process.

Visual formats have advanced beyond single-view images and videos: 3D movies are commonplace, researchers have developed multi-view navigation systems, and VR is helping to push light field cameras to mass market. However, editing tools for these media are still nascent, and even simple filtering operations like color correction or stylization are problematic: naively applying image filters per frame or per view rarely produces satisfying results due to time and space inconsistencies. Our method preserves and stabilizes filter effects while being agnostic to the inner working of the filter. It captures filter effects in the gradient domain, then uses input frame gradients as a reference to impose temporal and spatial consistency. Our least-squares formulation adds minimal overhead compared to naive data processing. Further, when filter cost is high, we introduce a filter transfer strategy that reduces the number of per-frame filtering computations by an order of magnitude, with only a small reduction in visual quality. We demonstrate our algorithm on several camera array formats including stereo videos, light fields, and wide baselines.

In this article, we present BubbleView, an alternative methodology for eye tracking using discrete mouse clicks to measure which information people consciously choose to examine. BubbleView is a mouse-contingent, moving-window interface in which participants are presented with a series of blurred images and click to reveal “bubbles” -- small, circular areas of the image at original resolution, similar to having a confined area of focus like the eye fovea. Across 10 experiments with 28 different parameter combinations, we evaluated BubbleView on a variety of image types: information visualizations, natural images, static webpages, and graphic designs, and compared the clicks to eye fixations collected with eye-trackers in controlled lab settings. We found that BubbleView clicks can both (i) successfully approximate eye fixations on different images, and (ii) be used to rank image and design elements by importance. BubbleView is designed to collect clicks on static images, and works best for defined tasks such as describing the content of an information visualization or measuring image importance. BubbleView data is cleaner and more consistent than related methodologies that use continuous mouse movements. Our analyses validate the use of mouse-contingent, moving-window methodologies as approximating eye fixations for different image and task types.

High-throughput and high-content screening enables large scale, cost-effective experiments in which cell cultures are exposed to a wide spectrum of drugs. The resulting multivariate data sets have a large but shallow hierarchical structure. The deepest level of this structure describes cells in terms of numeric features that are derived from image data. The subsequent level describes enveloping cell cultures in terms of imposed experiment conditions (exposure to drugs). We present Screenit, a visual analysis approach designed in close collaboration with screening experts. Screenit enables the navigation and analysis of multivariate data at multiple hierarchy levels and at multiple levels of detail. Screenit integrates the interactive modeling of cell physical states (phenotypes) and the effects of drugs on cell cultures (hits). In addition, quality control is enabled via the detection of anomalies that indicate low-quality data, while providing an interface that is designed to match workflows of screening experts. We demonstrate analyses for a real-world data set, CellMorph, with 6 million cells across 20,000 cell cultures.

In recent years, there is a growing need for communicating complex data in an accessible graphical form. Existing visualization creation tools support automatic visual encoding, but lack flexibility for creating custom design; on the other hand, freeform illustration tools require manual visual encoding, making the design process time-consuming and error-prone. In this paper, we present Data-Driven Guides (DDG), a technique for designing expressive information graphics in a graphic design environment. Instead of being confined by predefined templates or marks, designers can generate guides from data and use the guides to draw, place and measure custom shapes. We provide guides to encode data using three fundamental visual encoding channels: length, area, and position. Users can combine more than one guide to construct complex visual structures and map these structures to data. When underlying data is changed, we use a deformation technique to transform custom shapes using the guides as the backbone of the shapes. Our evaluation shows that data-driven guides allow users to create expressive and more accurate custom data-driven graphics.

Information hierarchies are difficult to express when real-world space or time constraints force traversing the hierarchy in linear presentations, such as in educational books and classroom courses. We present booc.io, which allows linear and non-linear presentation and navigation of educational concepts and material. To support a breadth of material for each concept, booc.io is Web based, which allows adding material such as lecture slides, book chapters, videos, and LTIs. A visual interface assists the creation of the needed hierarchical structures. The goals of our system were formed in expert interviews, and we explain how our design meets these goals. We adapt a real-world course into booc.io, and perform introductory qualitative evaluation with students.

We present a simple and effective blind image deblurring method based on the dark channel prior. Our work is inspired by the interesting observation that the dark channel of blurred images is less sparse. While most image patches in the clean image contain some dark pixels, these pixels are not dark when averaged with neighboring highintensity pixels during the blur process. This change in the sparsity of the dark channel is an inherent property of the blur process, which we both prove mathematically and validate using training data. Therefore, enforcing the sparsity of the dark channel helps blind deblurring on various scenarios, including natural, face, text, and low-illumination images. However, sparsity of the dark channel introduces a non-convex non-linear optimization problem. We introduce a linear approximation of the min operator to compute the dark channel. Our look-up-table-based method converges fast in practice and can be directly extended to non-uniform deblurring. Extensive experiments show that our method achieves state-of-the-art results on deblurring natural images and compares favorably methods that are well-engineered for specific scenarios.

The analysis of paths in graphs is highly relevant in many domains. Typically, path-related tasks are performed in node-link layouts. Unfortunately, graph layouts often do not scale to the size of many real world networks. Also, many networks are multivariate, i.e., contain rich attribute sets associated with the nodes and edges. These attributes are often critical in judging paths, but directly visualizing attributes in a graph layout exacerbates the scalability problem. In this paper, we present visual analysis solutions dedicated to path-related tasks in large and highly multivariate graphs. We show that by focusing on paths, we can address the scalability problem of multivariate graph visualization, equipping analysts with a powerful tool to explore large graphs. We introduce Pathfinder (Figure 1), a technique that provides visual methods to query paths, while considering various constraints. The resulting set of paths is visualized in both a ranked list and as a node-link diagram. For the paths in the list, we display rich attribute data associated with nodes and edges, and the node-link diagram provides topological context. The paths can be ranked based on topological properties, such as path length or average node degree, and scores derived from attribute data. Pathfinder is designed to scale to graphs with tens of thousands of nodes and edges by employing strategies such as incremental query results. We demonstrate Pathfinder's fitness for use in scenarios with data from a coauthor network and biological pathways.

We propose a novel approach to automated delineation of curvilinear structures that form complex and potentially loopy networks. By representing the image data as a graph of potential paths, we first show how to weight these paths using discriminatively-trained classifiers that are both robust and generic enough to be applied to very different imaging modalities. We then present an Integer Programming approach to finding the optimal subset of paths, subject to structural and topological constraints that eliminate implausible solutions. Unlike earlier approaches that assume a tree topology for the networks, ours explicitly models the fact that the networks may contain loops, and can reconstruct both cyclic and acyclic ones. We demonstrate the effectiveness of our approach on a variety of challenging datasets including aerial images of road networks and micrographs of neural arbors, and show that it outperforms state-of-the-art techniques.

Everyone, from a shopper buying shoes to a doctor palpating a growth, uses their sense of touch to learn about the world. 3D printing is a powerful technology because it gives us the ability to control the haptic impression an object creates. This is critical for both replicating existing, real-world constructs and designing novel ones. However, each 3D printer has different capabilities and supports different materials, leaving us to ask: How can we best replicate a given haptic result on a particular output device? In this work, we address the problem of mapping a real-world material to its nearest 3D printable counterpart by constructing a perceptual model for the compliance of nonlinearly elastic objects. We begin by building a perceptual space from experimentally obtained user comparisons of twelve 3D-printed metamaterials. By comparing this space to a number of hypothetical computational models, we identify those that can be used to accurately and efficiently evaluate human-perceived differences in nonlinear stiffness. Furthermore, we demonstrate how such models can be applied to complex geometries in an interaction-aware way where the compliance is influenced not only by the material properties from which the object is made but also its geometry. We demonstrate several applications of our method in the context of fabrication and evaluate them in a series of user experiments.

In the field of connectomics, neuroscientists acquire electron microscopy volumes at nanometer resolution in order to reconstruct a detailed wiring diagram of the neurons in the brain. The resulting image volumes, which often are hundreds of terabytes in size, need to be segmented to identify cell boundaries, synapses, and important cell organelles. However, the segmentation process of a single volume is very complex, time-intensive, and usually performed using a diverse set of tools and many users. To tackle the associated challenges, this paper presents NeuroBlocks, which is a novel visualization system for tracking the state, progress, and evolution of very large volumetric segmentation data in neuroscience. NeuroBlocks is a multi-user web-based application that seamlessly integrates the diverse set of tools that neuroscientists currently use for manual and semi-automatic segmentation, proofreading, visualization, and analysis. NeuroBlocks is the first system that integrates this heterogeneous tool set, providing crucial support for the management, provenance, accountability, and auditing of large-scale segmentations. We describe the design of NeuroBlocks, starting with an analysis of the domain-specific tasks, their inherent challenges, and our subsequent task abstraction and visual representation. We demonstrate the utility of our design based on two case studies that focus on different user roles and their respective requirements for performing and tracking the progress of segmentation and proofreading in a large real-world connectomics project.

In this paper we move beyond memorability and investigate how visualizations are recognized and recalled. For this study we labeled a dataset of 393 visualizations and analyzed the eye movements of 33 participants as well as thousands of participant-generated text descriptions of the visualizations. This allowed us to determine what components of a visualization attract people's attention, and what information is encoded into memory. Our findings quantitatively support many conventional qualitative design guidelines, including that (1) titles and supporting text should convey the message of a visualization, (2) if used appropriately, pictograms do not interfere with understanding and can improve recognition, and (3) redundancy helps effectively communicate the message. Importantly, we show that visualizations memorable "at-a-glance" are also capable of effectively conveying the message of the visualization. Thus, a memorable visualization is often also an effective one.

Alternative splicing is a process by which the same DNA sequence is used to assemble different proteins, called protein isoforms. Alternative splicing works by selectively omitting some of the coding regions (exons) typically associated with a gene. Detection of alternative splicing is difficult and uses a combination of advanced data acquisition methods and statistical inference. Knowledge about the abundance of isoforms is important for understanding both normal processes and diseases and to eventually improve treatment through targeted therapies. The data, however, is complex and current visualizations for isoforms are neither perceptually efficient nor scalable. To remedy this, we developed Vials, a novel visual analysis tool that enables analysts to explore the various datasets that scientists use to make judgments about isoforms: the abundance of reads associated with the coding regions of the gene, evidence for junctions, i.e., edges connecting the coding regions, and predictions of isoform frequencies. Vials is scalable as it allows for the simultaneous analysis of many samples in multiple groups. Our tool thus enables experts to (a) identify patterns of isoform abundance in groups of samples and (b) evaluate the quality of the data. We demonstrate the value of our tool in case studies using publicly available datasets.

Semi-automatic text analysis involves manual inspection of text. Often, different text annotations (like part-of-speech or named entities) are indicated by using distinctive text highlighting techniques. In typesetting there exist well-known formatting conventions, such as bold typeface, italics, or background coloring, that are useful for highlighting certain parts of a given text. Also, many advanced techniques for visualization and highlighting of text exist; yet, standard typesetting is common, and the effects of standard typesetting on the perception of text are not fully understood. As such, we surveyed and tested the effectiveness of common text highlighting techniques, both individually and in combination, to discover how to maximize pop-out effects while minimizing visual interference between techniques. To validate our findings, we conducted a series of crowdsourced experiments to determine: i) a ranking of nine commonly-used text highlighting techniques; ii) the degree of visual interference between pairs of text highlighting techniques; iii) the effectiveness of techniques for visual conjunctive search. Our results show that increasing font size works best as a single highlighting technique, and that there are significant visual interferences between some pairs of highlighting techniques. We discuss the pros and cons of different combinations as a design guideline to choose text highlighting techniques for text viewers.

Presents corrections for the paper, "Guidelines for effective usage of text highlighting techniques," (Strobelt, H., et al), IEEE Trans. Vis. Comput.Graph., vol. 22, no. 1, pp. 489-498, Jan. 2016.

In many learning tasks, the structure of the target space of a function holds rich information about the relationships between evaluations of functions on different data points. Existing approaches attempt to exploit this relationship information implicitly by enforcing smoothness on function evaluations only. However, what happens if we explicitly regularize the relationships between function evaluations? Inspired by homophily, we regularize based on a smooth relationship function, either defined from the data or with labels. In experiments, we demonstrate that this significantly improves the performance of state-of-the-art algorithms in semi-supervised classification and in spectral data embedding for constrained clustering and dimensionality reduction.

The common graph Laplacian regularizer is well-established in semi-supervised learning and spectral dimensionality reduction. However, as a first-order regularizer, it can lead to degenerate functions in high-dimensional manifolds. The iterated graph Laplacian enables high-order regularization, but it has a high computational complexity and so cannot be applied to large problems. We introduce a new regularizer which is globally high order and so does not suffer from the degeneracy of the graph Laplacian regularizer, but is also sparse for efficient computation in semi-supervised learning applications. We reduce computational complexity by building a local first-order approximation of the manifold as a surrogate geometry, and construct our high-order regularizer based on local derivative evaluations therein. Experiments on human body shape and pose analysis demonstrate the effectiveness and efficiency of our method.

As consumer depth sensors become widely available, estimating scene flow from RGBD sequences has received increasing attention. Although the depth information allows the recovery of 3D motion from a single view, it poses new challenges. In particular, depth boundaries are not well-aligned with RGB image edges and therefore not reliable cues to localize 2D motion boundaries. In addition, methods that extend the 2D optical flow formulation to 3D still produce large errors in occlusion regions. To better use depth for occlusion reasoning, we propose a layered RGBD scene flow method that jointly solves for the scene segmentation and the motion. Our key observation is that the noisy depth is sufficient to decide the depth ordering of layers, thereby avoiding a computational bottleneck for RGB layered methods. Furthermore, the depth enables us to estimate a per-layer 3D rigid motion to constrain the motion of each layer. Experimental results on both the Middlebury and real-world sequences demonstrate the effectiveness of the layered approach for RGBD scene flow estimation.

Existing approaches for diffusion on graphs, e.g., for label propagation, are mainly focused on isotropic diffusion, which is induced by the commonly-used graph Laplacian regularizer. Inspired by the success of diffusivity tensors for anisotropic diffusion in image processing, we presents anisotropic diffusion on graphs and the corresponding label propagation algorithm. We develop positive definite diffusivity operators on the vector bundles of Riemannian manifolds, and discretize them to diffusivity operators on graphs. This enables us to easily define new robust diffusivity operators which significantly improve semi-supervised learning performance over existing diffusion algorithms.

Light field displays allow viewers to see view-dependent 3D content as if looking through a window; however, existing work on light field display interaction is limited. Yet, they have the potential to parallel 2D pen and touch screen systems, which present a joint input and display surface for natural interaction. We propose a 4D display and interaction space using a dual-purpose lenslet array, which combines light field display and light field pen sensing, and allows us to estimate the 3D position and 2D orientation of the pen. This method is simple, fast (150Hz), with position accuracy of 2-3mm and precision of 0.2-0.6mm from 0-350mm away from the lenslet array, and orientation accuracy of 2 degrees and precision of 0.2-0.3 degrees within a 45 degree field of view. Further, we 3D print the lenslet array with embedded baffles to reduce out-of-bounds cross-talk, and use an optical relay to allow interaction behind the focal plane. We demonstrate our joint display/sensing system with interactive light field painting.

Abstract This survey gives an overview of the current state of the art in GPU techniques for interactive large‐scale volume visualization. Modern techniques in this field have brought about a sea change in how interactive visualization and analysis of giga‐, tera‐ and petabytes of volume data can be enabled on GPUs. In addition to combining the parallel processing power of GPUs with out‐of‐core methods and data streaming, a major enabler for interactivity is making both the computational and the visualization effort proportional to the amount and resolution of data that is actually visible on screen, i.e. ‘output‐sensitive’ algorithms and system designs. This leads to recent output‐sensitive approaches that are ‘ray‐guided’, ‘visualization‐driven’ or ‘display‐aware’. In this survey, we focus on these characteristics and propose a new categorization of GPU‐based large‐scale volume visualization techniques based on the notions of actual output‐resolution visibility and the current working set of volume bricks—the current subset of data that is minimally required to produce an output image of the desired display resolution. Furthermore, we discuss the differences and similarities of different rendering and data traversal strategies in volume rendering by putting them into a common context—the notion of address translation. For our purposes here, we view parallel (distributed) visualization using clusters as an orthogonal set of techniques that we do not discuss in detail but that can be used in conjunction with what we present in this survey.

Metallophones such as glockenspiels produce sounds in response to contact. Building these instruments is a complicated process, limiting their shapes to well-understood designs such as bars. We automatically optimize the shape of arbitrary 2D and 3D objects through deformation and perforation to produce sounds when struck which match user-supplied frequency and amplitude spectra. This optimization requires navigating a complex energy landscape, for which we develop Latin Complement Sampling to both speed up finding minima and provide probabilistic bounds on landscape exploration. Our method produces instruments which perform similarly to those that have been professionally-manufactured, while also expanding the scope of shape and sound that can be realized, e.g., single object chords. Furthermore, we can optimize sound spectra to create overtones and to dampen specific frequencies. Thus our technique allows even novices to design metallophones with unique sound and appearance.

Extending image processing techniques to videos is a non-trivial task; applying processing independently to each video frame often leads to temporal inconsistencies, and explicitly encoding temporal consistency requires algorithmic changes. We describe a more general approach to temporal consistency. We propose a gradient-domain technique that is blind to the particular image processing algorithm. Our technique takes a series of processed frames that suffers from flickering and generates a temporally-consistent video sequence. The core of our solution is to infer the temporal regularity from the original unprocessed video, and use it as a temporal consistency guide to stabilize the processed sequence. We formally characterize the frequency properties of our technique, and demonstrate, in practice, its ability to stabilize a wide range of popular image processing techniques including enhancement and stylization of color and tone, intrinsic images, and depth estimation.

Motion-tracked real-time character control is important for games and VR, but current solutions are limited: retargeting is hard for non-human characters, with locomotion bound to the sensing volume; and pose mappings are ambiguous with difficult dynamic motion control. We robustly estimate wave properties --- amplitude, frequency, and phase ---for a set of interactively-defined gestures by mapping user motions to a low-dimensional independent representation. The mapping separates simultaneous or intersecting gestures, and extrapolates gesture variations from single training examples. For animations such as locomotion, wave properties map naturally to stride length, step frequency, and progression, and allow smooth transitions from standing, to walking, to running. Interpolating out-of-phase locomotions is hard, e.g., quadruped legs between walks and runs switch phase, so we introduce a new time-interpolation scheme to reduce artifacts. These improvements to real-time motion-tracked character control are important for common cyclic animations. We validate this in a user study, and show versatility to apply to part- and full-body motions across a variety of sensors.

Most motion estimation algorithms (optical flow, layered models) cannot handle large amount of occlusion in textureless regions, as motion is often initialized with no occlusion assumption despite that occlusion may be included in the final objective. To handle such situations, we propose a local layering model where motion and occlusion relationships are inferred jointly. In particular, the uncertainties of occlusion relationships are retained so that motion is inferred by considering all the possibilities of local occlusion relationships. In addition, the local layering model handles articulated objects with self-occlusion. We demonstrate that the local layering model can handle motion and occlusion well for both challenging synthetic and real sequences.

Abstract This paper presents a technique to recover geometry from time‐lapse sequences of outdoor scenes. We build upon photometric stereo techniques to recover approximate shadowing, shading and normal components allowing us to alter the material and normals of the scene. Previous work in analyzing such images has faced two fundamental difficulties: 1. the illumination in outdoor images consists of time‐varying sunlight and skylight, and 2. the motion of the sun is restricted to a near‐planar arc through the sky, making surface normal recovery unstable. We develop methods to estimate the reflection component due to skylight illumination. We also show that sunlight directions are usually non‐planar, thus making surface normal recovery possible. This allows us to estimate approximate surface normals for outdoor scenes using a single day of data. We demonstrate the use of these surface normals for a number of image editing applications including reflectance, lighting, and normal editing.

The facial performance of an individual is inherently rich in subtle deformation and timing details. Although these subtleties make the performance realistic and compelling, they often elude both motion capture and hand animation. We present a technique for adding fine-scale details and expressiveness to low-resolution art-directed facial performances, such as those created manually using a rig, via marker-based capture, by fitting a morphable model to a video, or through Kinect reconstruction using recent faceshift technology. We employ a high-resolution facial performance capture system to acquire a representative performance of an individual in which he or she explores the full range of facial expressiveness. From the captured data, our system extracts an expressiveness model that encodes subtle spatial and temporal deformation details specific to that particular individual. Once this model has been built, these details can be transferred to low-resolution art-directed performances. We demonstrate results on various forms of input; after our enhancement, the resulting animations exhibit the same nuances and fine spatial details as the captured performance, with optional temporal enhancement to match the dynamics of the actor. Finally, we show that our technique outperforms the current state-of-the-art in example-based facial animation.

Separating a photograph into its reflectance and illumination intrinsic images is a fundamentally ambiguous problem, and state-of-the-art algorithms combine sophisticated reflectance and illumination priors with user annotations to create plausible results. However, these algorithms cannot be easily extended to videos for two reasons: first, näively applying algorithms designed for single images to videos produce results that are temporally incoherent; second, effectively specifying user annotations for a video requires interactive feedback, and current approaches are orders of magnitudes too slow to support this. We introduce a fast and temporally consistent algorithm to decompose video sequences into their reflectance and illumination components. Our algorithm uses a hybrid ℓ 2 ℓ p formulation that separates image gradients into smooth illumination and sparse reflectance gradients using look-up tables. We use a multi-scale parallelized solver to reconstruct the reflectance and illumination from these gradients while enforcing spatial and temporal reflectance constraints and user annotations. We demonstrate that our algorithm automatically produces reasonable results, that can be interactively refined by users, at rates that are two orders of magnitude faster than existing tools, to produce high-quality decompositions for challenging real-world video sequences. We also show how these decompositions can be used for a number of video editing applications including recoloring, retexturing, illumination editing, and lighting-aware compositing.

We present NeuroLines, a novel visualization technique designed for scalable detailed analysis of neuronal connectivity at the nanoscale level. The topology of 3D brain tissue data is abstracted into a multi-scale, relative distance-preserving subway map visualization that allows domain scientists to conduct an interactive analysis of neurons and their connectivity. Nanoscale connectomics aims at reverse-engineering the wiring of the brain. Reconstructing and analyzing the detailed connectivity of neurons and neurites (axons, dendrites) will be crucial for understanding the brain and its development and diseases. However, the enormous scale and complexity of nanoscale neuronal connectivity pose big challenges to existing visualization techniques in terms of scalability. NeuroLines offers a scalable visualization framework that can interactively render thousands of neurites, and that supports the detailed analysis of neuronal structures and their connectivity. We describe and analyze the design of NeuroLines based on two real-world use-cases of our collaborators in developmental neuroscience, and investigate its scalability to large-scale neuronal connectivity data.

As the size of image data from microscopes and telescopes increases, the need for high-throughput processing and visualization of large volumetric data has become more pressing. At the same time, many-core processors and GPU accelerators are commonplace, making high-performance distributed heterogeneous computing systems affordable. However, effectively utilizing GPU clusters is difficult for novice programmers, and even experienced programmers often fail to fully leverage the computing power of new parallel architectures due to their steep learning curve and programming complexity. In this paper, we propose Vivaldi, a new domain-specific language for volume processing and visualization on distributed heterogeneous computing systems. Vivaldi's Python-like grammar and parallel processing abstractions provide flexible programming tools for non-experts to easily write high-performance parallel computing code. Vivaldi provides commonly used functions and numerical operators for customized visualization and high-throughput image processing applications. We demonstrate the performance and usability of Vivaldi on several examples ranging from volume rendering to image segmentation.

Answering questions about complex issues often requires analysts to take into account information contained in multiple interconnected datasets. A common strategy in analyzing and visualizing large and heterogeneous data is dividing it into meaningful subsets. Interesting subsets can then be selected and the associated data and the relationships between the subsets visualized. However, neither the extraction and manipulation nor the comparison of subsets is well supported by state-of-the-art techniques. In this paper we present Domino, a novel multiform visualization technique for effectively representing subsets and the relationships between them. By providing comprehensive tools to arrange, combine, and extract subsets, Domino allows users to create both common visualization techniques and advanced visualizations tailored to specific use cases. In addition to the novel technique, we present an implementation that enables analysts to manage the wide range of options that our approach offers. Innovative interactive features such as placeholders and live previews support rapid creation of complex analysis setups. We introduce the technique and the implementation using a simple example and demonstrate scalability and effectiveness in a use case from the field of cancer genomics.

Proofreading refers to the manual correction of automatic segmentations of image data. In connectomics, electron microscopy data is acquired at nanometer-scale resolution and results in very large image volumes of brain tissue that require fully automatic segmentation algorithms to identify cell boundaries. However, these algorithms require hundreds of corrections per cubic micron of tissue. Even though this task is time consuming, it is fairly easy for humans to perform corrections through splitting, merging, and adjusting segments during proofreading. In this paper we present the design and implementation of Mojo, a fully-featured single-user desktop application for proofreading, and Dojo, a multi-user web-based application for collaborative proofreading. We evaluate the accuracy and speed of Mojo, Dojo, and Raveler, a proofreading tool from Janelia Farm, through a quantitative user study. We designed a between-subjects experiment and asked non-experts to proofread neurons in a publicly available connectomics dataset. Our results show a significant improvement of corrections using web-based Dojo, when given the same amount of time. In addition, all participants using Dojo reported better usability. We discuss our findings and provide an analysis of requirements for designing visual proofreading software.

Understanding relationships between sets is an important analysis task that has received widespread attention in the visualization community. The major challenge in this context is the combinatorial explosion of the number of set intersections if the number of sets exceeds a trivial threshold. In this paper we introduce UpSet, a novel visualization technique for the quantitative analysis of sets, their intersections, and aggregates of intersections. UpSet is focused on creating task-driven aggregates, communicating the size and properties of aggregates and intersections, and a duality between the visualization of the elements in a dataset and their set membership. UpSet visualizes set intersections in a matrix layout and introduces aggregates based on groupings and queries. The matrix layout enables the effective representation of associated data, such as the number of elements in the aggregates and intersections, as well as additional summary statistics derived from subset or element attributes. Sorting according to various measures enables a task-driven analysis of relevant intersections and aggregates. The elements represented in the sets and their associated attributes are visualized in a separate view. Queries based on containment in specific intersections, aggregates or driven by attribute filters are propagated between both views. We also introduce several advanced visual encodings and interaction methods to overcome the problems of varying scales and to address scalability. UpSet is web-based and open source. We demonstrate its general utility in multiple use cases from various domains.

Large scale data analysis is nowadays a crucial part of drug discovery. Biologists and chemists need to quickly explore and evaluate potentially effective yet safe compounds based on many datasets that are in relationship with each other. However, there is a lack of tools that support them in these processes. To remedy this, we developed ConTour, an interactive visual analytics technique that enables the exploration of these complex, multi-relational datasets. At its core ConTour lists all items of each dataset in a column. Relationships between the columns are revealed through interaction: selecting one or multiple items in one column highlights and re-sorts the items in other columns. Filters based on relationships enable drilling down into the large data space. To identify interesting items in the first place, ConTour employs advanced sorting strategies, including strategies based on connectivity strength and uniqueness, as well as sorting based on item attributes. ConTour also introduces interactive nesting of columns, a powerful method to show the related items of a child column for each item in the parent column. Within the columns, ConTour shows rich attribute data about the items as well as information about the connection strengths to other datasets. Finally, ConTour provides a number of detail views, which can show items from multiple datasets and their associated data at the same time. We demonstrate the utility of our system in case studies conducted with a team of chemical biologists, who investigate the effects of chemical compounds on cells and need to understand the underlying mechanisms.

Layered models allow scene segmentation and motion estimation to be formulated together and to inform one another. Traditional layered motion methods, however, employ fairly weak models of scene structure, relying on locally connected Ising/Potts models which have limited ability to capture long-range correlations in natural scenes. To address this, we formulate a fully-connected layered model that enables global reasoning about the complicated segmentations of real objects. Optimization with fully-connected graphical models is challenging, and our inference algorithm leverages recent work on efficient mean field updates for fully-connected conditional random fields. These methods can be implemented efficiently using high-dimensional Gaussian filtering. We combine these ideas with a layered flow model, and find that the long-range connections greatly improve segmentation into figure-ground layers when compared with locally connected MRF models. Experiments on several benchmark datasets show that the method can recover fine structures and large occlusion regions, with good flow accuracy and much lower computational cost than previous locally-connected layered models.

We propose a novel approach to automated delineation of linear structures that form complex and potentially loopy networks. This is in contrast to earlier approaches that usually assume a tree topology for the networks. At the heart of our method is an Integer Programming formulation that allows us to find the global optimum of an objective function designed to allow cycles but penalize spurious junctions and early terminations. We demonstrate that it outperforms state-of-the-art techniques on a wide range of datasets.

To study the emergence of cooperative behavior, we have developed a scalable parallel framework for evolutionary game dynamics. This is a critical computational tool enabling large-scale agent simulation research. An important aspect is the amount of history, or memory steps, that each agent can keep. When six memory steps are taken into account, the strategy space spans 24096 potential strategies, requiring large populations of agents. We introduce a multi-level decomposition method that allows us to exploit both multi-node and thread-level parallel scaling while minimizing communication overhead. We present the results of a production run modeling up to six memory steps for populations consisting of up to 1018 agents, making this study one of the largest yet undertaken. The high rate of mutation within the population results in a non-trivial parallel implementation. The strong and weak scaling studies provide insight into parallel scalability and programmability trade-offs for large-scale simulations, while exhibiting near perfect weak and strong scaling on 16,384 tasks on Blue Gene/Q. We further show 99% weak scaling up to 294,912 processors 82% strong scaling efficiency up to 262,144 processors of Blue Gene/P. Our framework marks an important step in the study of game dynamics with potential applications in fields ranging from biology to economics and sociology.

Abstract Collaborative slide image viewing systems are becoming increasingly important in pathology applications such as telepathology and E‐learning. Despite rapid advances in computing and imaging technology, current digital pathology systems have limited performance with respect to remote viewing of whole slide images on desktop or mobile computing devices. In this paper we present a novel digital pathology client–server system that supports collaborative viewing of multi‐plane whole slide images over standard networks using multi‐touch‐enabled clients. Our system is built upon a standard HTTP web server and a MySQL database to allow multiple clients to exchange image and metadata concurrently. We introduce a domain‐specific image‐stack compression method that leverages real‐time hardware decoding on mobile devices. It adaptively encodes image stacks in a decorrelated colour space to achieve extremely low bitrates (0.8 bpp) with very low loss of image quality. We evaluate the image quality of our compression method and the performance of our system for diagnosis with an in‐depth user study.

In most professional cinema productions, the color palette of the movie is painstakingly adjusted by a team of skilled colorists -- through a process referred to as color grading -- to achieve a certain visual look. The time and expertise required to grade a video makes it difficult for amateurs to manipulate the colors of their own video clips. In this work, we present a method that allows a user to transfer the color palette of a model video clip to their own video sequence. We estimate a per-frame color transform that maps the color distributions in the input video sequence to that of the model video clip. Applying this transformation naively leads to artifacts such as bleeding and flickering. Instead, we propose a novel differential-geometry-based scheme that interpolates these transformations in a manner that minimizes their curvature, similarly to curvature flows. In addition, we automatically determine a set of keyframes that best represent this interpolated transformation curve, and can be used subsequently, to manually refine the color grade. We show how our method can successfully transfer color palettes between videos for a range of visual styles and a number of input video clips.

This paper presents ConnectomeExplorer, an application for the interactive exploration and query-guided visual analysis of large volumetric electron microscopy (EM) data sets in connectomics research. Our system incorporates a knowledge-based query algebra that supports the interactive specification of dynamically evaluated queries, which enable neuroscientists to pose and answer domain-specific questions in an intuitive manner. Queries are built step by step in a visual query builder, building more complex queries from combinations of simpler queries. Our application is based on a scalable volume visualization framework that scales to multiple volumes of several teravoxels each, enabling the concurrent visualization and querying of the original EM volume, additional segmentation volumes, neuronal connectivity, and additional meta data comprising a variety of neuronal data attributes. We evaluate our application on a data set of roughly one terabyte of EM data and 750 GB of segmentation data, containing over 4,000 segmented structures and 1,000 synapses. We demonstrate typical use-case scenarios of our collaborators in neuroscience, where our system has enabled them to answer specific scientific questions using interactive querying and analysis on the full-size data for the first time.

An ongoing debate in the Visualization community concerns the role that visualization types play in data understanding. In human cognition, understanding and memorability are intertwined. As a first step towards being able to ask questions about impact and effectiveness, here we ask: 'What makes a visualization memorable?' We ran the largest scale visualization study to date using 2,070 single-panel visualizations, categorized with visualization type (e.g., bar chart, line graph, etc.), collected from news media sites, government reports, scientific journals, and infographic sources. Each visualization was annotated with additional attributes, including ratings for data-ink ratios and visual densities. Using Amazon's Mechanical Turk, we collected memorability scores for hundreds of these visualizations, and discovered that observers are consistent in which visualizations they find memorable and forgettable. We find intuitive results (e.g., attributes like color and the inclusion of a human recognizable object enhance memorability) and less intuitive results (e.g., common graphs are less memorable than unique visualization types). Altogether our findings suggest that quantifying memorability is a general metric of the utility of information, an essential step towards determining how to design effective visualizations.

Having effective visualizations of filesystem provenance data is valuable for understanding its complex hierarchical structure. The most common visual representation of provenance data is the node-link diagram. While effective for understanding local activity, the node-link diagram fails to offer a high-level summary of activity and inter-relationships within the data. We present a new tool, InProv, which displays filesystem provenance with an interactive radial-based tree layout. The tool also utilizes a new time-based hierarchical node grouping method for filesystem provenance data we developed to match the user's mental model and make data exploration more intuitive. We compared InProv to a conventional node-link based tool, Orbiter, in a quantitative evaluation with real users of filesystem provenance data including provenance data experts, IT professionals, and computational scientists. We also compared in the evaluation our new node grouping method to a conventional method. The results demonstrate that InProv results in higher accuracy in identifying system activity than Orbiter with large complex data sets. The results also show that our new time-based hierarchical node grouping method improves performance in both tools, and participants found both tools significantly easier to use with the new time-based node grouping method. Subjective measures show that participants found InProv to require less mental activity, less physical activity, less work, and is less stressful to use. Our study also reveals one of the first cases of gender differences in visualization; both genders had comparable performance with InProv, but women had a significantly lower average accuracy (56%) compared to men (70%) with Orbiter.

Rankings are a popular and universal approach to structuring otherwise unorganized collections of items by computing a rank for each item based on the value of one or more of its attributes. This allows us, for example, to prioritize tasks or to evaluate the performance of products relative to each other. While the visualization of a ranking itself is straightforward, its interpretation is not, because the rank of an item represents only a summary of a potentially complicated relationship between its attributes and those of the other items. It is also common that alternative rankings exist which need to be compared and analyzed to gain insight into how multiple heterogeneous attributes affect the rankings. Advanced visual exploration tools are needed to make this process efficient. In this paper we present a comprehensive analysis of requirements for the visualization of multi-attribute rankings. Based on these considerations, we propose LineUp--a novel and scalable visualization technique that uses bar charts. This interactive technique supports the ranking of items based on multiple heterogeneous attributes with different scales and semantics. It enables users to interactively combine attributes and flexibly refine parameters to explore the effect of changes in the attribute combination. This process can be employed to derive actionable insights as to which attributes of an item need to be modified in order for its rank to change. Additionally, through integration of slope graphs, LineUp can also be used to compare multiple alternative rankings on the same set of items, for example, over time or across different attribute combinations. We evaluate the effectiveness of the proposed multi-attribute visualization technique in a qualitative study. The study shows that users are able to successfully solve complex ranking tasks in a short period of time.

Biological pathway maps are highly relevant tools for many tasks in molecular biology. They reduce the complexity of the overall biological network by partitioning it into smaller manageable parts. While this reduction of complexity is their biggest strength, it is, at the same time, their biggest weakness. By removing what is deemed not important for the primary function of the pathway, biologists lose the ability to follow and understand cross-talks between pathways. Considering these cross-talks is, however, critical in many analysis scenarios, such as judging effects of drugs. In this paper we introduce Entourage, a novel visualization technique that provides contextual information lost due to the artificial partitioning of the biological network, but at the same time limits the presented information to what is relevant to the analyst's task. We use one pathway map as the focus of an analysis and allow a larger set of contextual pathways. For these context pathways we only show the contextual subsets, i.e., the parts of the graph that are relevant to a selection. Entourage suggests related pathways based on similarities and highlights parts of a pathway that are interesting in terms of mapped experimental data. We visualize interdependencies between pathways using stubs of visual links, which we found effective yet not obtrusive. By combining this approach with visualization of experimental data, we can provide domain experts with a highly valuable tool. We demonstrate the utility of Entourage with case studies conducted with a biochemist who researches the effects of drugs on pathways. We show that the technique is well suited to investigate interdependencies between pathways and to analyze, understand, and predict the effect that drugs have on different cell types.

Abstract Close collaboration with other scientific fields is an important goal for the visualization community. Yet engaging in a scientific collaboration can be challenging. The physical sciences, namely astronomy, chemistry, earth sciences and physics, exhibit an extensive range of research directions, providing exciting challenges for visualization scientists and creating ample possibilities for collaboration. We present the first survey of its kind that provides a comprehensive view of existing work on visualization for the physical sciences. We introduce novel classification schemes based on application area, data dimensionality and main challenge addressed, and apply these classifications to each contribution from the literature. Our survey helps in understanding the status of current research and serves as a useful starting point for those interested in visualization for the physical sciences.

In this paper, we present a novel image classification system that is built around a core of trainable filter ensembles that we call Volterra kernel classifiers. Our system treats images as a collection of possibly overlapping patches and is composed of three components: (1) A scheme for a single patch classification that seeks a smooth, possibly nonlinear, functional mapping of the patches into a range space, where patches of the same class are close to one another, while patches from different classes are far apart-in the L_2 sense. This mapping is accomplished using trainable convolution filters (or Volterra kernels) where the convolution kernel can be of any shape or order. (2) Given a corpus of Volterra classifiers with various kernel orders and shapes for each patch, a boosting scheme for automatically selecting the best weighted combination of the classifiers to achieve higher per-patch classification rate. (3) A scheme for aggregating the classification information obtained for each patch via voting for the parent image classification. We demonstrate the effectiveness of the proposed technique using face recognition as an application area and provide extensive experiments on the Yale, CMU PIE, Extended Yale B, Multi-PIE, and MERL Dome benchmark face data sets. We call the Volterra kernel classifiers applied to face recognition Volterrafaces. We show that our technique, which falls into the broad class of embedding-based face image discrimination methods, consistently outperforms various state-of-the-art methods in the same category.

Articulated deformable characters are widespread in computer animation. Unfortunately, we lack methods for their automatic fabrication using modern additive manufacturing (AM) technologies. We propose a method that takes a skinned mesh as input, then estimates a fabricatable single-material model that approximates the 3D kinematics of the corresponding virtual articulated character in a piecewise linear manner. We first extract a set of potential joint locations. From this set, together with optional, user-specified range constraints, we then estimate mechanical friction joints that satisfy inter-joint non-penetration and other fabrication constraints. To avoid brittle joint designs, we place joint centers on an approximate medial axis representation of the input geometry, and maximize each joint's minimal cross-sectional area. We provide several demonstrations, manufactured as single, assembled pieces using 3D printers.

This paper presents the first volume visualization system that scales to petascale volumes imaged as a continuous stream of high-resolution electron microscopy images. Our architecture scales to dense, anisotropic petascale volumes because it: (1) decouples construction of the 3D multi-resolution representation required for visualization from data acquisition, and (2) decouples sample access time during ray-casting from the size of the multi-resolution hierarchy. Our system is designed around a scalable multi-resolution virtual memory architecture that handles missing data naturally, does not pre-compute any 3D multi-resolution representation such as an octree, and can accept a constant stream of 2D image tiles from the microscopes. A novelty of our system design is that it is visualization-driven: we restrict most computations to the visible volume data. Leveraging the virtual memory architecture, missing data are detected during volume ray-casting as cache misses, which are propagated backwards for on-demand out-of-core processing. 3D blocks of volume data are only constructed from 2D microscope image tiles when they have actually been accessed during ray-casting. We extensively evaluate our system design choices with respect to scalability and performance, compare to previous best-of-breed systems, and illustrate the effectiveness of our system for real microscopy data from neuroscience.

We describe a unified framework for generating a single high-quality still image ("snapshot") from a short video clip. Our system allows the user to specify the desired operations for creating the output image, such as super resolution, noise and blur reduction, and selection of best focus. It also provides a visual summary of activity in the video by incorporating saliency-based objectives in the snapshot formation process. We show examples on a number of different video clips to illustrate the utility and flexibility of our system.

We describe a set of image editing and viewing tools that explicitly take into account the resolution of the display on which the image is viewed. Our approach is twofold. First, we design editing tools that process only the visible data, which is useful for images larger than the display. This encompasses cases such as multi-image panoramas and high-resolution medical data. Second, we propose an adaptive way to set viewing parameters such brightness and contrast. Because we deal with very large images, different locations and scales often require different viewing parameters. We let users set these parameters at a few places and interpolate satisfying values everywhere else. We demonstrate the efficiency of our approach on different display and image sizes. Since the computational complexity to render a view depends on the display resolution and not the actual input image resolution, we achieve interactive image editing even on a 16 gigapixel image.

We present two theses in this paper: First, performance of most existing face recognition algorithms improves if instead of the whole image, smaller patches are individually classified followed by label aggregation using voting. Second, weighted plurality1voting outperforms other popular voting methods if the weights are set such that they maximize the victory margin for the winner with respect to each of the losers. Moreover, this can be done while taking higher order relationships among patches into account using kernels. We call this scheme Kernel Plurality. We verify our proposals with detailed experimental results and show that our framework with Kernel Plurality improves the performance of various face recognition algorithms beyond what has been previously reported in the literature. Furthermore, on five different benchmark datasets - Yale A, CMU PIE, MERL Dome, Extended Yale B and Multi-PIE, we show that Kernel Plurality in conjunction with recent face recognition algorithms can provide state-of-the-art results in terms of face recognition rates.

We address the problem of automatic 3D segmentation of a stack of electron microscopy sections of brain tissue. Unlike previous efforts, where the reconstruction is usually done on a section-to-section basis, or by the agglomerative clustering of 2D segments, we leverage information from the entire volume to obtain a globally optimal 3D segmentation. To do this, we formulate the segmentation as the solution to a fusion problem. We first enumerate multiple possible 2D segmentations for each section in the stack, and a set of 3D links that may connect segments across consecutive sections. We then identify the fusion of segments and links that provide the most globally consistent segmentation of the stack. We show that this two-step approach of pre-enumeration and posterior fusion yields significant advantages and provides state-of-the-art reconstruction results. Finally, as part of this method, we also introduce a robust rotationally-invariant set of features that we use to learn and enumerate the above 2D segmentations. Our features outperform previous connectomic-specific descriptors without relying on a large set of heuristics or manually designed filter banks.

We present a method for replacing facial performances in video. Our approach accounts for differences in identity, visual appearance, speech, and timing between source and target videos. Unlike prior work, it does not require substantial manual operation or complex acquisition hardware, only single-camera video. We use a 3D multilinear model to track the facial performance in both videos. Using the corresponding 3D geometry, we warp the source to the target face and retime the source to match the target performance. We then compute an optimal seam through the video volume that maintains temporal consistency in the final composite. We showcase the use of our method on a variety of examples and present the result of a user study that suggests our results are difficult to distinguish from real video footage.

Heart disease is the number one killer in the United States, and finding indicators of the disease at an early stage is critical for treatment and prevention. In this paper we evaluate visualization techniques that enable the diagnosis of coronary artery disease. A key physical quantity of medical interest is endothelial shear stress (ESS). Low ESS has been associated with sites of lesion formation and rapid progression of disease in the coronary arteries. Having effective visualizations of a patient's ESS data is vital for the quick and thorough non-invasive evaluation by a cardiologist. We present a task taxonomy for hemodynamics based on a formative user study with domain experts. Based on the results of this study we developed HemoVis, an interactive visualization application for heart disease diagnosis that uses a novel 2D tree diagram representation of coronary artery trees. We present the results of a formal quantitative user study with domain experts that evaluates the effect of 2D versus 3D artery representations and of color maps on identifying regions of low ESS. We show statistically significant results demonstrating that our 2D visualizations are more accurate and efficient than 3D representations, and that a perceptually appropriate color map leads to fewer diagnostic mistakes than a rainbow color map.

Computer-generated (CG) images have achieved high levels of realism. This realism, however, comes at the cost of long and expensive manual modeling, and often humans can still distinguish between CG and real images. We introduce a new data-driven approach for rendering realistic imagery that uses a large collection of photographs gathered from online repositories. Given a CG image, we retrieve a small number of real images with similar global structure. We identify corresponding regions between the CG and real images using a mean-shift cosegmentation algorithm. The user can then automatically transfer color, tone, and texture from matching regions to the CG image. Our system only uses image processing operations and does not require a 3D model of the scene, making it fast and easy to integrate into digital content creation workflows. Results of a user study show that our hybrid images appear more realistic than the originals.

Abstract Biologists pioneering the new field of comparative functional genomics attempt to infer the mechanisms of gene regulation by looking for similarities and differences of gene activity over time across multiple species. They use three kinds of data: functional data such as gene activity measurements, pathway data that represent a series of reactions within a cellular process, and phylogenetic relationship data that describe the relatedness of species. No existing visualization tool can visually encode the biologically interesting relationships between multiple pathways, multiple genes, and multiple species. We tackle the challenge of visualizing all aspects of this comparative functional genomics dataset with a new interactive tool called Pathline. In addition to the overall characterization of the problem and design of Pathline, our contributions include two new visual encoding techniques. One is a new method for linearizing metabolic pathways that provides appropriate topological information and supports the comparison of quantitative data along the pathway. The second is the curvemap view, a depiction of time series data for comparison of gene activity and metabolite levels across multiple species. Pathline was developed in close collaboration with a team of genomic scientists. We validate our approach with case studies of the biologists’ use of Pathline and report on how they use the tool to confirm existing findings and to discover new scientific insights.

This paper introduces a data-driven process for designing and fabricating materials with desired deformation behavior. Our process starts with measuring deformation properties of base materials. For each base material we acquire a set of example deformations, and we represent the material as a non-linear stress-strain relationship in a finite-element model. We have validated our material measurement process by comparing simulations of arbitrary stacks of base materials with measured deformations of fabricated material stacks. After material measurement, our process continues with designing stacked layers of base materials. We introduce an optimization process that finds the best combination of stacked layers that meets a user's criteria specified by example deformations. Our algorithm employs a number of strategies to prune poor solutions from the combinatorial search space. We demonstrate the complete process by designing and fabricating objects with complex heterogeneous materials using modern multi-material 3D printers.

We investigate a complete pipeline for measuring, modeling, and fabricating objects with specified subsurface scattering behaviors. The process starts with measuring the scattering properties of a given set of base materials, determining their radial reflection and transmission profiles. We describe a mathematical model that predicts the profiles of different stackings of base materials, at arbitrary thicknesses. In an inverse process, we can then specify a desired reflection profile and compute a layered composite material that best approximates it. Our algorithm efficiently searches the space of possible combinations of base materials, pruning unsatisfactory states imposed by physical constraints. We validate our process by producing both homogeneous and heterogeneous composites fabricated using a multi-material 3D printer. We demonstrate reproductions that have scattering properties approximating complex materials.

Traditional image compositing techniques, such as alpha matting and gradient domain compositing, are used to create composites that have plausible boundaries. But when applied to images taken from different sources or shot under different conditions, these techniques can produce unrealistic results. In this work, we present a framework that explicitly matches the visual appearance of images through a process we call image harmonization , before blending them. At the heart of this framework is a multi-scale technique that allows us to transfer the appearance of one image to another. We show that by carefully manipulating the scales of a pyramid decomposition of an image, we can match contrast, texture, noise, and blur, while avoiding image artifacts. The output composite can then be reconstructed from the modified pyramid coefficients while enforcing both alpha-based and seamless boundary constraints. We show how the proposed framework can be used to produce realistic composites with minimal user interaction in a number of different scenarios.

These are the proceedings of the IEEE Visualization Conference 2010 (Vis 2010) and the IEEE Information Visualization Conference 2010 (InfoVis 2010) held during October 24 to 29, 2010 in Salt Lake City, Utah, USA.

Histology is the study of the structure of biological tissue using microscopy techniques. As digital imaging technology advances, high resolution microscopy of large tissue volumes is becoming feasible; however, new interactive tools are needed to explore and analyze the enormous datasets. In this paper we present a visualization framework that specifically targets interactive examination of arbitrarily large image stacks. Our framework is built upon two core techniques: display-aware processing and GPU-accelerated texture compression. With display-aware processing, only the currently visible image tiles are fetched and aligned on-the-fly, reducing memory bandwidth and minimizing the need for time-consuming global pre-processing. Our novel texture compression scheme for GPUs is tailored for quick browsing of image stacks. We evaluate the usability of our viewer for two histology applications: digital pathology and visualization of neural structure at nanoscale-resolution in serial electron micrographs.

Cells in an organism share the same genetic information in their DNA, but have very different forms and behavior because of the selective expression of subsets of their genes. The widely used approach of measuring gene expression over time from a tissue sample using techniques such as microarrays or sequencing do not provide information about the spatial position within the tissue where these genes are expressed. In contrast, we are working with biologists who use techniques that measure gene expression in every individual cell of entire fruitfly embryos over an hour of their development, and do so for multiple closely-related subspecies of Drosophila. These scientists are faced with the challenge of integrating temporal gene expression data with the spatial location of cells and, moreover, comparing this data across multiple related species. We have worked with these biologists over the past two years to develop MulteeSum, a visualization system that supports inspection and curation of data sets showing gene expression over time, in conjunction with the spatial location of the cells where the genes are expressed--it is the first tool to support comparisons across multiple such data sets. MulteeSum is part of a general and flexible framework we developed with our collaborators that is built around multiple summaries for each cell, allowing the biologists to explore the results of computations that mix spatial information, gene expression measurements over time, and data from multiple related species or organisms. We justify our design decisions based on specific descriptions of the analysis needs of our collaborators, and provide anecdotal evidence of the efficacy of MulteeSum through a series of case studies.

The ability to constrain the geometry of deformable models for image segmentation can be useful when information about the expected shape or positioning of the objects in a scene is known a priori. An example of this occurs when segmenting neural cross sections in electron microscopy. Such images often contain multiple nested boundaries separating regions of homogeneous intensities. For these applications, multiphase level sets provide a partitioning framework that allows for the segmentation of multiple deformable objects by combining several level set functions. Although there has been much effort in the study of statistical shape priors that can be used to constrain the geometry of each partition, none of these methods allow for the direct modeling of geometric arrangements of partitions. In this paper, we show how to define elastic couplings between multiple level set functions to model ribbon-like partitions. We build such couplings using dynamic force fields that can depend on the image content and relative location and shape of the level set functions. To the best of our knowledge, this is the first work that shows a direct way of geometrically constraining multiphase level sets for image segmentation. We demonstrate the robustness of our method by comparing it with previous level set segmentation methods.

We present an image restoration method that leverages a large database of images gathered from the web. Given an input image, we execute an efficient visual search to find the closest images in the database; these images define the input's visual context. We use the visual context as an image-specific prior and show its value in a variety of image restoration operations, including white balance correction, exposure correction, and contrast enhancement. We evaluate our approach using a database of 1 million images downloaded from Flickr and demonstrate the effect of database size on performance. Our results show that priors based on the visual context consistently out-perform generic or even domain-specific priors for these operations.

This paper introduces a data-driven representation and modeling technique for simulating non-linear heterogeneous soft tissue. It simplifies the construction of convincing deformable models by avoiding complex selection and tuning of physical material parameters, yet retaining the richness of non-linear heterogeneous behavior. We acquire a set of example deformations of a real object, and represent each of them as a spatially varying stress-strain relationship in a finite-element model. We then model the material by non-linear interpolation of these stress-strain relationships in strain-space. Our method relies on a simple-to-build capture system and an efficient run-time simulation algorithm based on incremental loading, making it suitable for interactive computer graphics applications. We present the results of our approach for several non-linear materials and biological soft tissue, with accurate agreement of our model to the measured data.

Recent advances in scanning technology provide high resolution EM (Electron Microscopy) datasets that allow neuro-scientists to reconstruct complex neural connections in a nervous system. However, due to the enormous size and complexity of the resulting data, segmentation and visualization of neural processes in EM data is usually a difficult and very time-consuming task. In this paper, we present NeuroTrace, a novel EM volume segmentation and visualization system that consists of two parts: a semi-automatic multiphase level set segmentation with 3D tracking for reconstruction of neural processes, and a specialized volume rendering approach for visualization of EM volumes. It employs view-dependent on-demand filtering and evaluation of a local histogram edge metric, as well as on-the-fly interpolation and ray-casting of implicit surfaces for segmented neural structures. Both methods are implemented on the GPU for interactive performance. NeuroTrace is designed to be scalable to large datasets and data-parallel hardware architectures. A comparison of NeuroTrace with a commonly used manual EM segmentation tool shows that our interactive workflow is faster and easier to use for the reconstruction of complex neural processes.

In the field of comparative genomics, scientists seek to answer questions about evolution and genomic function by comparing the genomes of species to find regions of shared sequences. Conserved syntenic blocks are an important biological data abstraction for indicating regions of shared sequences. The goal of this work is to show multiple types of relationships at multiple scales in a way that is visually comprehensible in accordance with known perceptual principles. We present a task analysis for this domain where the fundamental questions asked by biologists can be understood by a characterization of relationships into the four types of proximity/location, size, orientation, and similarity/strength, and the four scales of genome, chromosome, block, and genomic feature. We also propose a new taxonomy of the design space for visually encoding conservation data. We present MizBee, a multiscale synteny browser with the unique property of providing interactive side-by-side views of the data across the range of scales supporting exploration of all of these relationship types. We conclude with case studies from two biologists who used MizBee to augment their previous automatic analysis work flow, providing anecdotal evidence about the efficacy oft he system for the visualization of syntenic data, the analysis of conservation relationships, and the communication of scientific insights.

Automultiscopic displays show 3D stereoscopic images that can be viewed from any viewpoint without special glasses. These displays are becoming widely available and affordable. In this paper, we describe how an automultiscopic display, built for viewing 3D images, can be repurposed to display 2D interfaces that appear differently from different points-of-view. For single-user applications, point-of-view becomes a means of input and a user is able to reveal different views of an application by simply moving their head left and right. For multi-user applications, a single-display application can show each member of the group a different variation of the interface. We outline three types of multi-view interfaces and illustrate each with example applications.

We present a real-time algorithm to estimate the 3D pose of a previously unseen face from a single range image. Based on a novel shape signature to identify noses in range images, we generate candidates for their positions, and then generate and evaluate many pose hypotheses in parallel using modern graphics processing units (GPUs). We developed a novel error function that compares the input range image to precomputed pose images of an average face model. The algorithm is robust to large pose variations of plusmn90deg yaw, plusmn45deg pitch and plusmn30deg roll rotation, facial expression, partial occlusion, and works for multiple faces in the field of view. It correctly estimates 97.8% of the poses within yaw and pitch error of 15deg at 55.8 fps. To evaluate the algorithm, we built a database of range images with large pose variations and developed a method for automatic ground truth annotation.

In an extended image sequence of an outdoor scene, one observes changes in color induced by variations in the spectral composition of daylight. This paper proposes a model for these temporal color changes and explores its use for the analysis of outdoor scenes from time-lapse video data. We show that the time-varying changes in direct sunlight and ambient skylight can be recovered with this model, and that an image sequence can be decomposed into two corresponding components. The decomposition provides access to both radiometric and geometric information about a scene, and we demonstrate how this can be exploited for a variety of visual tasks, including color-constancy, background subtraction, shadow detection, scene reconstruction, and camera geo-location.

The method of Moving Least Squares (MLS) is a popular framework for reconstructing continuous functions from scattered data due to its rich mathematical properties and well-understood theoretical foundations. This paper applies MLS to volume rendering, providing a unified mathematical framework for ray casting of scalar data stored over regular as well as irregular grids. We use the MLS reconstruction to render smooth isosurfaces and to compute accurate derivatives for high-quality shading effects. We also present a novel, adaptive preintegration scheme to improve the efficiency of the ray casting algorithm by reducing the overall number of function evaluations, and an efficient implementation of our framework exploiting modern graphics hardware. The resulting system enables high-quality volume integration and shaded isosurface rendering for regular and irregular volume data.

Methods that faithfully and robustly capture the geometry of complex material interfaces in labeled volume data are important for generating realistic and accurate visualizations and simulations of real-world objects. The generation of such multimaterial models from measured data poses two unique challenges: first, the surfaces must be well-sampled with regular, efficient tessellations that are consistent across material boundaries; and second, the resulting meshes must respect the nonmanifold geometry of the multimaterial interfaces. This paper proposes a strategy for sampling and meshing multimaterial volumes using dynamic particle systems, including a novel, differentiable representation of the material junctions that allows the particle system to explicitly sample corners, edges, and surfaces of material intersections. The distributions of particles are controlled by fundamental sampling constraints, allowing Delaunay-based meshing algorithms to reliably extract watertight meshes of consistently high-quality.

The proliferation of low-cost infrared cameras gives us a new angle for attacking many unsolved vision problems by leveraging a larger range of the electromagnetic spectrum. A first step to utilizing these images is to explore the statistics of infrared images and compare them to the corresponding statistics in the visible spectrum. In this paper, we analyze the power spectra as well as the marginal and joint wavelet coefficient distributions of datasets of indoor and outdoor images. We note that infrared images have noticeably less texture indoors where temperatures are more homogenous. The joint wavelet statistics also show strong correlation between object boundaries in IR and visible images, leading to high potential for vision applications using a combined statistical model.

This paper addresses signal processing issues related to coded representation, reconstruction and rendering of multiview video for 3D displays. We provide an overview of standardization efforts for multiview video that are aimed at reducing data rates required to represent the multiview video in compressed form. We then present an anti-aliasing filtering technique that effectively eliminates ghosting artifacts when rendering multiview video on 3D displays. Since high-frequency components of the signal are removed, substantial reductions in the compressed data rate could also be realized. Finally, we discuss the importance of scalability in the context of multiview video coding and suggest a combined anti-aliasing and scalable decoding scheme to minimize decoding resources for a given 3D display.

Multi-view 3D displays are preferable to other stereoscopic display technologies because they provide autostereoscopic viewing from any viewpoint without special glasses. However, they require a large number of pixels to achieve high image quality. Therefore, data compression is a major issue for this approach. In this paper, we present a framework for efficient compression of multi-view video streams for multi-view 3D displays. Our goal is to optimize image quality without increasing the required data bandwidth. We achieve this by taking into account a precise notion of the multi-dimensional display bandwidth. The display bandwidth implies that scene elements that appear at a given distance from the display become increasingly blurry as the distance grows. Our main contribution is to enhance conventional multi-view compression pipelines with an additional pre-filtering step that bandlimits the multi-view signal to the display bandwidth. This imposes a shallow depth of field on the input images, thereby removing high frequency content. We show that this pre-filtering step leads to increased image quality compared to state-of-the-art multi-view coding at equal bitrate. We present results of an extensive user study that corroborate the benefits of our approach. Our work suggests that display pre-filtering will be a fundamental component in signal processing for 3D displays, and that any multi-view compression scheme will benefit from our pre-filtering technique.

We present a novel multi-scale representation and acquisition method for the animation of high-resolution facial geometry and wrinkles. We first acquire a static scan of the face including reflectance data at the highest possible quality. We then augment a traditional marker-based facial motion-capture system by two synchronized video cameras to track expression wrinkles. The resulting model consists of high-resolution geometry, motion-capture data, and expression wrinkles in 2D parametric form. This combination represents the facial shape and its salient features at multiple scales. During motion synthesis the motion-capture data deforms the high-resolution geometry using a linear shell-based mesh-deformation method. The wrinkle geometry is added to the facial base mesh using nonlinear energy optimization. We present the results of our approach for performance replay as well as for wrinkle editing.

We describe a method for converting time-lapse photography captured with outdoor cameras into Factored Time-Lapse Video (FTLV): a video in which time appears to move faster (i.e., lapsing) and where data at each pixel has been factored into shadow, illumination, and reflectance components. The factorization allows a user to easily relight the scene, recover a portion of the scene geometry (normals), and to perform advanced image editing operations. Our method is easy to implement, robust, and provides a compact representation with good reconstruction characteristics. We show results using several publicly available time-lapse sequences.

As GPU-powered special effects become more sophisticated, it becomes harder to create and manage effect interaction using the fairly primitive shading languages. This difficulty also introduces a workflow problem: artists design effects but only programmers can implement them, making it impossible for them to work asynchronously.To address these problems we present abstract shade trees and heuristic algorithms that operate over them. The trees allow designers to easily create effects by connecting primitives such as cube mapping and modulation. These primitives publish semantically rich types that encapsulate notions like vector basis and normalization. The algorithms employ these published types to automatically infer atomic and compound connectors between the primitives, and generate code for the tree. We also describe a visual editing environment for specifying the trees.Our data structure and algorithms spare designers from having to specify low-level programming details, enabling them to experiment without depending on programmers. The algorithms ensure that the generated code will be free of type-mismatches, a problem in previous shade trees. The abstract shade tree can also naturally express high-level features like shadows and reflections whose implementations overlap; that cross-cutting has made them difficult to modularize in more traditional ways. In experiments, the generated shaders are as efficient as handwritten code.

Detailed surface geometry contributes greatly to the visual realism of 3D face models. However, acquiring high-resolution face geometry is often tedious and expensive. Consequently, most face models used in games, virtual reality, or computer vision look unrealistically smooth. In this paper, we introduce a new statistical technique for the analysis and synthesis of small three-dimensional facial features, such as wrinkles and pores. We acquire high-resolution face geometry for people across a wide range of ages, genders, and races. For each scan, we separate the skin surface details from a smooth base mesh using displaced subdivision surfaces. Then, we analyze the resulting displacement maps using the texture analysis/synthesis framework of Heeger and Bergen, adapted to capture statistics that vary spatially across a face. Finally, we use the extracted statistics to synthesize plausible detail on face meshes of arbitrary subjects. We demonstrate the effectiveness of this method in several applications, including analysis of facial texture in subjects with different ages and genders, interpolation between high-resolution face scans, adding detail to low-resolution face scans, and adjusting the apparent age of faces. In all cases, we are able to re-produce fine geometric details consistent with those observed in high resolution scans.

Recent progress in the measurement of surface reflectance has created a demand for non-parametric appearance representations that are accurate, compact, and easy to use for rendering. Another crucial goal, which has so far received little attention, is editability: for practical use, we must be able to change both the directional and spatial behavior of surface reflectance (e.g., making one material shinier, another more anisotropic, and changing the spatial "texture maps" indicating where each material appears). We introduce an Inverse Shade Tree framework that provides a general approach to estimating the "leaves" of a user-specified shade tree from high-dimensional measured datasets of appearance. These leaves are sampled 1- and 2-dimensional functions that capture both the directional behavior of individual materials and their spatial mixing patterns. In order to compute these shade trees automatically, we map the problem to matrix factorization and introduce a flexible new algorithm that allows for constraints such as non-negativity, sparsity, and energy conservation. Although we cannot infer every type of shade tree, we demonstrate the ability to reduce multi-gigabyte measured datasets of the Spatially-Varying Bidirectional Reflectance Distribution Function (SVBRDF) into a compact representation that may be edited in real time.

We have measured 3D face geometry, skin reflectance, and subsurface scattering using custom-built devices for 149 subjects of varying age, gender, and race. We developed a novel skin reflectance model whose parameters can be estimated from measurements. The model decomposes the large amount of measured skin data into a spatially-varying analytic BRDF, a diffuse albedo map, and diffuse subsurface scattering. Our model is intuitive, physically plausible, and -- since we do not use the original measured data -- easy to edit as well. High-quality renderings come close to reproducing real photographs. The analysis of the model parameters for our sample population reveals variations according to subject age, gender, skin type, and external factors (e.g., sweat, cold, or makeup). Using our statistics, a user can edit the overall appearance of a face (e.g., changing skin type and age) or change small-scale features using texture synthesis (e.g., adding moles and freckles). We are making the collected statistics publicly available to the research community for applications in face synthesis and analysis.

We present a technique to generate an illumination subspace for arbitrary 3D faces based on the statistics of measured illuminations under variable lighting conditions from many subjects. A bilinear model based on the higher-order singular value decomposition is used to create a compact illumination subspace given arbitrary shape parameters from a parametric 3D face model. Using a fitting procedure based on minimizing the distance of the input image to the dynamically changing illumination subspace, we reconstruct a shape-specific illumination subspace from a single photograph. We use the reconstructed illumination subspace in various face recognition experiments with variable lighting conditions and obtain accuracies which are very competitive with previous methods that require specific training sessions or multiple images of the subject.

We present a system for three-dimensional visualization of complex liquid chromatography-mass spectrometry (LCMS) data. Every LCMS data point has three attributes: time, mass, and intensity. Instead of the traditional visualization of two-dimensional subsets of the data, we visualize it as a height field or terrain in 3D. Unlike traditional terrains, LCMS data has non-linear sampling and consists mainly of tall needle-like features. We adapt the level-of-detail techniques of geometry clipmaps for hardware-accelerated rendering of LCMS data. The data is cached in video memory as a set of nested rectilinear grids centered about the view frustum. We introduce a simple compression scheme and dynamically stream data from the CPU to the GPU as the viewpoint moves. Our system allows interactive investigation of complex LCMS data with close to one billion data points at up to 130 frames per second, depending on the view conditions.

Video matting is the process of pulling a high-quality alpha matte and foreground from a video sequence. Current techniques require either a known background (e.g., a blue screen) or extensive user interaction (e.g., to specify known foreground and background elements). The matting problem is generally under-constrained, since not enough information has been collected at capture time. We propose a novel, fully autonomous method for pulling a matte using multiple synchronized video streams that share a point of view but differ in their plane of focus. The solution is obtained by directly minimizing the error in filter-based image formation equations, which are over-constrained by our rich data stream. Our system solves the fully dynamic video matting problem without user assistance: both the foreground and background may be high frequency and have dynamic content, the foreground may resemble the background, and the scene is lit by natural (as opposed to polarized or collimated) illumination.

We present a vision-based performance interface for controlling animated human characters. The system interactively combines information about the user's motion contained in silhouettes from three viewpoints with domain knowledge contained in a motion capture database to produce an animation of high quality. Such an interactive system might be useful for authoring, for teleconferencing, or as a control interface for a character in a game. In our implementation, the user performs in front of three video cameras; the resulting silhouettes are used to estimate his orientation and body configuration based on a set of discriminative local features. Those features are selected by a machine-learning algorithm during a preprocessing step. Sequences of motions that approximate the user's actions are extracted from the motion database and scaled in time to match the speed of the user's motion. We use swing dancing, a complex human motion, to demonstrate the effectiveness of our approach. We compare our results to those obtained with a set of global features, Hu moments, and ground truth measurements from a motion capture system.

Face Transfer is a method for mapping videorecorded performances of one individual to facial animations of another. It extracts visemes (speech-related mouth articulations), expressions, and three-dimensional (3D) pose from monocular video or film footage. These parameters are then used to generate and drive a detailed 3D textured face mesh for a target identity, which can be seamlessly rendered back into target footage. The underlying face model automatically adjusts for how the target performs facial expressions and visemes. The performance data can be easily edited to change the visemes, expressions, pose, or even the identity of the target---the attributes are separably controllable. This supports a wide variety of video rewrite and puppetry applications.Face Transfer is based on a multilinear model of 3D face meshes that separably parameterizes the space of geometric variations due to different attributes (e.g., identity, expression, and viseme). Separability means that each of these attributes can be independently varied. A multilinear model can be estimated from a Cartesian product of examples (identities × expressions × visemes) with techniques from statistical analysis, but only after careful preprocessing of the geometric data set to secure one-to-one correspondence, to minimize cross-coupling artifacts, and to fill in any missing examples. Face Transfer offers new solutions to these problems and links the estimated model with a face-tracking algorithm to extract pose, expression, and viseme parameters.

Animation of photorealistic computer graphics models is an important goal for many applications. Image-based modeling has emerged as a promising approach to capture and visualize real-world objects. Animating image-based models, however, is still a largely unsolved problem. In this paper, we extend a popular image-based representation called surface reflectance field to animate and render deformable real-world objects under arbitrary illumination. Deforming the surface reflectance field is achieved by modifying the underlying impostor geometry. We augment the impostor by a local parameterization that allows the correct evaluation of acquired reflectance images, preserving the original light model on the deformed surface. We present a deferred shading scheme to handle the increased amount of data involved in shading the deformable surface reflectance field. We show animations of various objects that were acquired with 3D photography.

In this paper, we present a coding framework addressing image-space compression for free-viewpoint video. Our framework is based on time-varying 3D point samples which represent real-world objects. The 3D point samples are obtained after a geometrical reconstruction from multiple pre-recorded video sequences and thus allow for arbitrary viewpoints during playback. The encoding of the data is performed as an off-line process and is not time-critical. The decoding however, must support for real-time rendering of the dynamic 3D data. We introduce a compression framework which encodes multiple point attributes like depth and color into progressive streams. The reference data structure is aligned on the original camera input images and thus enables for easy view-dependent decoding. A novel differential coding approach permits random access in constant time throughout the entire data set and thus enables arbitrary viewpoint trajectories in both time and space.

We present a hardware-accelerated adaptive EWA (elliptical weighted average) volume splatting algorithm. EWA splatting combines a Gaussian reconstruction kernel with a low-pass image filter for high image quality without aliasing artifacts or excessive blurring. We introduce a novel adaptive filtering scheme to reduce the computational cost of EWA splatting. We show how this algorithm can be efficiently implemented on modern graphics processing units (GPUs). Our implementation includes interactive classification and fast lighting. To accelerate the rendering we store splat geometry and 3D volume data locally in GPU memory. We present results for several rectilinear volume datasets that demonstrate the high image quality and interactive rendering speed of our method.

Advances in graphics hardware and rendering methods are shaping the future of visualization. For example, programmable graphics processors are redefining the traditional visualization cycle. In some cases it is now possible to run the computational simulation and associated visualization side-by-side on the same chip. Moreover, global illumination and non-photorealistic effects promise to deliver imagery which enables greater insight into high resolution, multivariate, and higher-dimensional data. The panelists will offer distinct viewpoints on the direction of future graphics hardware and its potential impact on visualization, and on the nature of advanced visualizationrelated tools and techniques. Presentation of these viewpoints will be followed by audience participation in the form of a question and answer period moderated by the panel organizer.

Three-dimensional TV is expected to be the next revolution in the history of television. We implemented a 3D TV prototype system with real-time acquisition, transmission, and 3D display of dynamic scenes. We developed a distributed, scalable architecture to manage the high computation and bandwidth demands. Our system consists of an array of cameras, clusters of network-connected PCs, and a multi-projector 3D display. Multiple video streams are individually encoded and sent over a broadband network to the display. The 3D display shows high-resolution (1024 × 768) stereoscopic color images for multiple viewpoints without special glasses. We implemented systems with rear-projection and front-projection lenticular screens. In this paper, we provide a detailed overview of our 3D TV system, including an examination of design choices and tradeoffs. We present the calibration and image alignment procedures that are necessary to achieve good image quality. We present qualitative results and some early user feedback. We believe this is the first real-time end-to-end 3D TV system with enough views and resolution to provide a truly immersive 3D experience.

Summary form only given. The field of computer graphics abounds with modeling and simulation problems. Among these are the representation of surface shape, the description of surface reflectance, the probabilistic modeling of small-scale variations, and the application of physics for simulating the dynamics of rigid and elastic materials. During its formative years, computer graphics has focused largely on developing algorithms and systems for performing efficient simulations that transform these analytic representations into images and animations. At present, the simulation framework for computer graphics is very mature. In last ten years, we have also witnessed significant technological developments in the areas of high-quality sensors and measurement devices. However, the data provided from these devices are frequently incompatible with the representations assumed by most computer graphics systems. The author explores new approaches to computer graphics that attempt to bridge the dichotomy between parametric and empirical modeling.

We present new hardware-accelerated techniques for rendering surface light fields with opacity hulls that allow for interactive visualization of objects that have complex reflectance properties and elaborate geometrical details. The opacity hull is a shape enclosing the object with view-dependent opacity parameterized onto that shape. We call the combination of opacity hulls and surface light fields the opacity light field. Opacity light fields are ideally suited for rendering of the visually complex objects and scenes obtained with 3D photography. We show how to implement opacity light fields in the framework of three surface light field rendering methods: view-dependent texture mapping, unstructured lumigraph rendering, and light field mapping. The modified algorithms can be effectively supported on modern graphics hardware. Our results show that all three implementations are able to achieve interactive or real-time frame rates.

We present a generative model for isotropic bidirectional reflectance distribution functions (BRDFs) based on acquired reflectance data. Instead of using analytical reflectance models, we represent each BRDF as a dense set of measurements. This allows us to interpolate and extrapolate in the space of acquired BRDFs to create new BRDFs. We treat each acquired BRDF as a single high-dimensional vector taken from a space of all possible BRDFs. We apply both linear (subspace) and non-linear (manifold) dimensionality reduction tools in an effort to discover a lower-dimensional representation that characterizes our measurements. We let users define perceptually meaningful parametrization directions to navigate in the reduced-dimension BRDF space. On the low-dimensional manifold, movement along these directions produces novel but valid BRDFs.

Elliptical weighted average (EWA) surface splatting is a technique for high quality rendering of point-sampled 3D objects. EWA surface splatting renders water-tight surfaces of complex point models with high quality, anisotropic texture filtering. In this paper we introduce a new multi-pass approach to perform EWA surface splatting on modern PC graphics hardware, called object space EWA splatting. We derive an object space formulation of the EWA filter, which is amenable for acceleration by conventional triangle-based graphics hardware. We describe how to implement the object space EWA filter using a two pass rendering algorithm. In the first rendering pass, visibility splatting is performed by shifting opaque surfel polygons backward along the viewing rays, while in the second rendering pass view-dependent EWA prefiltering is performed by deforming texture mapped surfel polygons. We use texture mapping and alpha blending to facilitate the splatting process. We implement our algorithm using programmable vertex and pixel shaders, fully exploiting the capabilities of today's graphics processing units (GPUs). Our implementation renders up to 3 million points per second on recent PC graphics hardware, an order of magnitude more than a pure software implementation of screen space EWA surface splatting. Categories and Subject Descriptors (according to ACM CCS): I.3.3 [Computer Graphics]: Display Algorithms

We have built a system for acquiring and displaying high quality graphical models of objects that are impossible to scan with traditional scanners. Our system can acquire highly specular and fuzzy materials, such as fur and feathers. The hardware set-up consists of a turntable, two plasma displays, an array of cameras, and a rotating array of directional lights. We use multi-background matting techniques to acquire alpha mattes of the object from multiple viewpoints. The alpha mattes are used to construct an opacity hull. The opacity hull is a new shape representation, defined as the visual hull of the object with view-dependent opacity. It enables visualization of complex object silhouettes and seamless blending of objects into new environments. Our system also supports relighting of objects with arbitrary appearance using surface reflectance fields, a purely image-based appearance representation. Our system is the first to acquire and render surface reflectance fields under varying illumination from arbitrary viewpoints. We have built three generations of digitizers with increasing sophistication. In this paper, we present our results from digitizing hundreds of models.

We present a framework for high quality splatting based on elliptical Gaussian kernels. To avoid aliasing artifacts, we introduce the concept of a resampling filter, combining a reconstruction kernel with a low-pass filter. Because of the similarity to Heckbert's (1989) EWA (elliptical weighted average) filter for texture mapping, we call our technique EWA splatting. Our framework allows us to derive EWA splat primitives for volume data and for point-sampled surface data. It provides high image quality without aliasing artifacts or excessive blurring for volume data and, additionally, features anisotropic texture filtering for point-sampled surfaces. It also handles nonspherical volume kernels efficiently; hence, it is suitable for regular, rectilinear, and irregular volume datasets. Moreover, our framework introduces a novel approach to compute the footprint function, facilitating efficient perspective projection of arbitrary elliptical kernels at very little additional cost. Finally, we show that EWA volume reconstruction kernels can be reduced to surface reconstruction kernels. This makes our splat primitive universal in rendering surface and volume data.

Modern laser range and optical scanners need rendering techniques that can handle millions of points with high resolution textures. This paper describes a point rendering and texture filtering technique called surface splatting which directly renders opaque and transparent surfaces from point clouds without connectivity. It is based on a novel screen space formulation of the Elliptical Weighted Average (EWA) filter. Our rigorous mathematical analysis extends the texture resampling framework of Heckbert to irregularly spaced point samples. To render the points, we develop a surface splat primitive that implements the screen space EWA filter. Moreover, we show how to optimally sample image and procedural textures to irregular point data during pre-processing. We also compare the optimal algorithm with a more efficient view-independent EWA pre-filter. Surface splatting makes the benefits of EWA texture filtering available to point-based rendering. It provides high quality anisotropic texture filtering, hidden surface removal, edge anti-aliasing, and order-independent transparency.

In this paper we present a novel framework for direct volume rendering using a splatting approach based on elliptical Gaussian kernels. To avoid aliasing artifacts, we introduce the concept of a resampling filter combining a reconstruction with a low-pass kernel. Because of the similarity to Heckbert's EWA (elliptical weighted average) filter for texture mapping we call our technique EWA volume splatting. It provides high image quality without aliasing artifacts or excessive blurring even with non-spherical kernels. Hence it is suitable for regular, rectilinear, and irregular volume data sets. Moreover, our framework introduces a novel approach to compute the footprint function. It facilitates efficient perspective projection of arbitrary elliptical kernels at very little additional cost. Finally, we show that EWA volume reconstruction kernels can be reduced to surface reconstruction kernels. This makes our splat primitive universal in reconstructing surface and volume data.

Surface elements (surfels) are a powerful paradigm to efficiently render complex geometric objects at interactive frame rates. Unlike classical surface discretizations, i.e., triangles or quadrilateral meshes, surfels are point primitives without explicit connectivity. Surfel attributes comprise depth, texture color, normal, and others. As a pre-process, an octree-based surfel representation of a geometric object is computed. During sampling, surfel positions and normals are optionally perturbed, and different levels of texture colors are prefiltered and stored per surfel. During rendering, a hierarchical forward warping algorithm projects surfels to a z-buffer. A novel method called visibility splatting determines visible surfels and holes in the z-buffer. Visible surfels are shaded using texture filtering, Phong illumination, and environment mapping using per-surfel normals. Several methods of image reconstruction, including supersampling, offer flexible speed-quality tradeoffs. Due to the simplicity of the operations, the surfel rendering pipeline is amenable for hardware implementation. Surfel objects offer complex shape, low rendering cost and high image quality, which makes them specifically suited for low-cost, real-time graphics, such as games.

(Article begins on next page) The Harvard community has made this article openly available. Please share how this access benefits you. Your story matters.

Real-time visualization of large volume data sets demands high-performance computation, pushing the storage, processing and data communication requirements to the limits of current technology. General-purpose parallel processors have been used to visualize moderate-size data sets at interactive frame rates; however, the cost and size of these supercomputers inhibits the widespread use for real-time visualization. This paper surveys several special-purpose architectures that seek to render volumes at interactive rates. These specialized visualization accelerators have cost, performance and size advantages over parallel processors. All architectures implement ray casting using parallel and pipelined hardware. We introduce a new metric that normalizes performance to compare these architectures. The architectures included in this survey are VOGUE, VIRIM, Array-Based Ray Casting, EM-Cube and VIZARD II. We also discuss future applications of special-purpose accelerators.

Article Design galleries: a general approach to setting parameters for computer graphics and animation Share on Authors: J. Marks MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , B. Andalman Harvard Univ. Harvard Univ.View Profile , P. A. Beardsley MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , W. Freeman MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , S. Gibson MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , J. Hodgins Georgia Tech. Georgia Tech.View Profile , T. Kang CMU CMUView Profile , B. Mirtich MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , H. Pfister MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , W. Ruml MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MA MERL - A Mitsubishi Electric Research Laboratory, 201 Broadway, Cambridge, MAView Profile , K. Ryall Harvard Univ. Harvard Univ.View Profile , J. Seims Univ. of Washington Univ. of WashingtonView Profile , S. Shieber Harvard Univ. Harvard Univ.View Profile Authors Info & Claims SIGGRAPH '97: Proceedings of the 24th annual conference on Computer graphics and interactive techniquesAugust 1997 Pages 389–400https://doi.org/10.1145/258734.258887Online:03 August 1997Publication History 346citation2,992DownloadsMetricsTotal Citations346Total Downloads2,992Last 12 Months156Last 6 weeks16 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Volumetric data sets require enormous storage capacity even at moderate resolution levels. The excessive storage demands not only stress the capacity of the underlying storage and communications systems, but also seriously limit the speed of volume rendering due to data movement and manipulation. A novel volumetric data visualization scheme is proposed and implemented in this work that renders 2D images directly from compressed 3D data sets. The novelty of this algorithm is that rendering is performed on the compressed representation of the volumetric data without pre-decompression. As a result, the overheads associated with both data movement and rendering processing are significantly reduced. The proposed algorithm generalizes previously proposed whole-volume frequency-domain rendering schemes by first dividing the 3D data set into subcubes, transforming each subcube to a frequency-domain representation, and applying the Fourier projection theorem to produce the projected 2D images according to given viewing angles. Compared to the whole-volume approach, the subcube-based scheme not only achieves higher compression efficiency by exploiting local coherency, but also improves the quality of resultant rendering images because it approximates the occlusion effect on a subcube by subcube basis.

This paper presents a novel approach to assist the user in exploring appropriate transfer functions for the visualization of volumetric datasets. The search for a transfer function is treated as a parameter optimization problem and addressed with stochastic search techniques. Starting from an initial population of (random or pre-defined) transfer functions, the evolution of the stochastic algorithms is controlled by either direct user selection of intermediate images or automatic fitness evaluation using user-specified objective functions. This approach essentially shields the user from the complex and tedious "trial and error" approach, and demonstrates effective and convenient generation of transfer functions.

Abstract Volume rendering is a key technique in scientific visualization that lends itself to significant exploitable parallelism. The high computational demands of real‐time volume rendering and continued technological advances in the area of VLSl give impetus to the development of special‐purpose volume rendering architectures. This paper presents and characterizes three recently developed volume rendering engines which are based on the ray‐casting method. A taxonomy of the algorithmic variants of ray‐casting and details of each ray‐casting architecture are discussed. The paper then compares the machinefeatures and provides an outlook onfuture developments in the area of volume rendering hardware.

VolVis is a diversified, easy to use, extensible, high performance, and portable volume visualization system for scientists and engineers as well as for visualization developers and researchers. VolVis accepts as input 3D scalar volumetric data as well as 3D volume-sampled and classical geometric models. Interaction with the data is controlled by a variety of 3D input devices in an input device-independent environment. VolVis output includes navigation preview, static images, and animation sequences. A variety of volume rendering algorithms are supported ranging from fast rough approximations, to compression-domain rendering, to accurate volumetric ray tracing and radiosity, and irregular grid rendering.>