In sparse novel view synthesis with few input views and wide baselines, existing methods often fail due to weak geometric correspondences and view-dependent color inconsistencies. Splatting-based approaches can produce plausible results near training views, but they frequently overfit and struggle to maintain smooth, realistic appearance transitions in novel viewpoints. We introduce a splat-based gradient-domain fusion method that addresses these limitations. Our approach first establishes reliable dense geometry via two-view stereo for stable initialization. We then generate intermediate virtual views by reprojecting input images, which provide reference gradient fields for gradient-domain fusion. By blending these gradients, our method transfers low-frequency, view-dependent colors to the rendered Gaussians, producing seamless appearance transitions across views. Extensive experiments show that our approach consistently outperforms state-of-the-art sparse Gaussian splatting methods, delivering robust and perceptually plausible view synthesis. A comprehensive user study further confirms that our results are perceptually preferred, with significantly smoother and more realistic color transitions than existing methods.

We present a method to reconstruct dynamic scenes from monocular continuous-wave time-of-flight (C-ToF) cameras using raw sensor samples that achieves similar or better accuracy than neural volumetric approaches and is 100× faster. Quickly achieving high-fidelity dynamic 3D reconstruction from a single viewpoint is a significant challenge in computer vision. In C-ToF radiance field reconstruction, the property of interest—depth—is not directly measured, causing an additional challenge. This problem has a large and underappreciated impact upon the optimization when using a fast primitive-based scene representation like 3D Gaussian splatting, which is commonly used with multi-view data to produce satisfactory results and is brittle in its optimization otherwise. We incorporate two heuristics into the optimization to improve the accuracy of scene geometry represented by Gaussians. Experimental results show that our approach produces accurate reconstructions under constrained C-ToF sensing conditions, including for fast motions like swinging baseball bats. https://visual.cs.brown.edu/gftorf

Large models have shown generalization across datasets for many low-level vision tasks, like depth estimation, but no such general models exist for scene flow. Even though scene flow prediction has wide potential, its practical use is limited because of the lack of generalization of current predictive models. We identify three key challenges and propose solutions for each. First, we create a method that jointly estimates geometry and motion for accurate prediction. Second, we alleviate scene flow data scarcity with a data recipe that affords us 1M annotated training samples across diverse synthetic scenes. Third, we evaluate different parameterizations for scene flow prediction and adopt a natural and effective parameterization. Our model outperforms existing methods as well as baselines built on large-scale models in terms of 3D end-point error, and shows zero-shot generalization to the casually captured videos from DAVIS and the robotic manipulation scenes from RoboTAP. Overall, our approach makes scene flow prediction more practical in-the-wild.Website: https://research.nvidia.com/labs/lpr/zeromsf/

In 3D object reconstruction from photographs, estimating material properties is challenging. We propose an inverse rendering method that uses active area lighting: as this provides a wider range of lighting angles per photo than point lighting, material reconstruction can be more accurate for the same number of photos. We compare area light shading with point lighting. With either mesh or 3D Gaussian splatting pipelines, area lighting can improve BRDF reconstruction and leads to +3 dB relighting PSNR over point lights, or need only \(\nicefrac {1}{5}\) of the input photos for the same quality. We also compare area light shading with Monte Carlo ray tracing and with differential linearly transformed cosines (LTC) plus shadow visibility weighting. LTC can be faster, improving optimization times by 25%. In SOTA method-level comparisons, our approach improves material reconstruction, particularly for material roughness, leading to superior relighting quality.

We propose a method that achieves state-of-the-art rendering quality and efficiency on monocular dynamic scene reconstruction using deformable 3D Gaussians. Implicit deformable representations commonly model motion with a canonical space and time-dependent backward-warping deformation field. Our method, GauFRe, uses a forward-warping deformation to explicitly model non-rigid transformations of scene geometry. Specifically, we propose a template set of 3D Gaussians residing in a canonical space, and a time-dependent forward-warping deformation field to model dynamic objects. Additionally, we tailor a 3D Gaussian-specific static component supported by an inductive bias-aware initialization approach which allows the deformation field to focus on moving scene regions, improving the rendering of complex real-world motion. The differentiable pipeline is optimized end-to-end with a self-supervised rendering loss. Experiments show our method achieves competitive results and higher efficiency than both previous state-of-the-art NeRF and Gaussian-based methods. For real-world scenes, GauFRe can train in ≈20 mins and offer 96 FPS real-time rendering on an RTX 3090 GPU.

We present a method to reconstruct indoor and outdoor static scene geometry and appearance from an omnidirectional video moving in a small circular sweep. This setting is challenging because of the small baseline and large depth ranges, making it difficult to find ray crossings. To better constrain the optimization, we estimate geometry as a signed distance field within a spherical binoctree data structure and use a complementary efficient tree traversal strategy based on a breadth-first search for sampling. Unlike regular grids or trees, the shape of this structure well-matches the camera setting, creating a better memory-quality trade-off. From an initial depth estimate, the binoc-tree is adaptively subdivided throughout the optimization, previous methods use a fixed depth that leaves the scene undersampled. In comparison with three neural optimization methods and two non-neural methods, ours shows de-creased geometry error on average, especially in a detailed scene, while significantly reducing the required number of voxels to represent such details

There is a widely-spread claim that GANs are difficult to train, and GAN architectures in the literature are littered with empirical tricks. We provide evidence against this claim and build a modern GAN baseline in a more principled manner. First, we derive a well-behaved regularized relativistic GAN loss that addresses issues of mode dropping and non-convergence that were previously tackled via a bag of ad-hoc tricks. We analyze our loss mathematically and prove that it admits local convergence guarantees, unlike most existing relativistic losses. Second, this loss allows us to discard all ad-hoc tricks and replace outdated backbones used in common GANs with modern architectures. Using StyleGAN2 as an example, we present a roadmap of simplification and modernization that results in a new minimalist baseline---R3GAN. Despite being simple, our approach surpasses StyleGAN2 on FFHQ, ImageNet, CIFAR, and Stacked MNIST datasets, and compares favorably against state-of-the-art GANs and diffusion models. Code: https://www.github.com/brownvc/R3GAN

PSNR 29.47 PSNR 28.80 PSNR 27.28 PSNR 30.84 PSNR 26.69 PSNR 26.

Augmented reality (AR) area labels can visualize real world regions with arbitrary boundaries and show invisible objects or features. But environment conditions such as lighting and clutter can decrease fixed or passive label visibility, and labels that have high opacity levels can occlude crucial details in the environment. We design and evaluate active AR area label visualization modes to enhance visibility across real-life environments, while still retaining environment details within the label. For this, we define a distant characteristic color from the environment in perceptual CIELAB space, then introduce spatial variations among label pixel colors based on the underlying environment variation. In a user study with 18 participants, we found that our active label visualization modes can be comparable in visibility to a fixed green baseline by Gabbard et al., and can outperform it with added spatial variation in cluttered environments, across varying levels of lighting (e.g., nighttime), and in environments with colors similar to the fixed baseline color.

3DMM conditioned face generation has gained traction due to its well-defined controllability; however, the trade-off is lower sample quality: Previous works such as DiscoFace-GAN and 3D-FM GAN show a significant FID gap compared to the unconditional StyleGAN, suggesting that there is a quality tax to pay for controllability. In this paper, we challenge the assumption that quality and controllability cannot coexist. To pinpoint the previous issues, we mathematically formalize the problem of 3DMM conditioned face generation. Then, we devise simple solutions to the problem under our proposed framework. This results in a new model that effectively removes the quality tax between 3DMM conditioned face GANs and the unconditional StyleGAN.Project webpage: visual.cs.brown.edu/taxfreegan

Lighting is crucial for portrait photography, yet the complex interactions between the skin and incident light are expensive to model computationally in graphics and difficult to reconstruct analytically via computer vision. Alternatively, to allow fast and controllable reflectance and lighting editing, we developed a physically based decomposition through deep learned priors from path-traced portrait images. Previous approaches that used simplified material models or low-frequency or low-dynamic-range lighting struggled to model specular reflections or relight directly without intermediate decomposition. However, we estimate the surface normal, skin albedo and roughness, and high-frequency HDRI maps, and propose an architecture to estimate both diffuse and specular reflectance components. In our experiments, we show that this approach can represent the true appearance function more effectively than simpler baseline methods, leading to better generalization and higher-quality editing.

How deep neural networks can aid visualization perception research is a wide-open question. This paper provides insights from three perspectives—prediction, generalization, and interpretation—via training and analyzing deep convolutional neural networks on human correlation judgments in scatterplots across three studies. The first study assesses the accuracy of twenty-nine neural network architectures in predicting human judgments, finding that a subset of the architectures (e.g., VGG-19) has comparable accuracy to the best-performing regression analyses in prior research. The second study shows that the resulting models from the first study display better generalizability than prior models on two other judgment datasets for different scatterplot designs. The third study interprets visual features learned by a convolutional neural network model, providing insights about how the model makes predictions, and identifies potential features that could be investigated in human correlation perception studies. Together, this paper suggests that deep neural networks can serve as a tool for visualization perception researchers in devising potential empirical study designs and hypothesizing about perpetual judgments. The preprint, data, code, and training logs are available at https://doi.org/10.17605/osf.io/exa8m.

From video, we reconstruct a neural volume that captures time-varying color, density, scene flow, semantics, and attention information. The semantics and attention let us identify salient foreground objects separately from the background across spacetime. To mitigate low resolution semantic and attention features, we compute pyramids that trade detail with whole-image context. After optimization, we perform a saliency-aware clustering to decompose the scene. To evaluate real-world scenes, we annotate object masks in the NVIDIA Dynamic Scene and DyCheck datasets. We demonstrate that this method can decompose dynamic scenes in an unsupervised way with competitive performance to a supervised method, and that it improves foreground/background segmentation over recent static/dynamic split methods. Project webpage: https://visual.cs.brown.edu/saff

We present an image formation model and optimization procedure that combines the advantages of neural radiance fields and structured light imaging. Existing depth-supervised neural models rely on depth sensors to accurately capture the scene’s geometry. However, the depth maps recovered by these sensors can be prone to error, or even fail outright. Instead of depending on the fidelity of processed depth maps from a structured light system, a more principled approach is to explicitly model the raw structured light images themselves. Our proposed approach enables the estimation of high-fidelity depth maps, including for objects with complex material properties (e.g., partially-transparent surfaces). Besides computing depth, the raw structured light images also confer other useful radiometric cues, which enable predicting surface normals and decomposing scene appearance in terms of a direct, indirect, and ambient component. We evaluate our framework quantitatively and qualitatively on a range of real and synthetic scenes, and decompose scenes into their constituent components for novel views.

Depth estimation is fundamental to 3D perception, and humans are known to have biased estimates of depth. This study investigates whether convolutional neural networks (CNNs) can be biased when predicting the sign of curvature and depth of surfaces of textured surfaces under different viewing conditions (field of view) and surface parameters (slant and texture irregularity). This hypothesis is drawn from the idea that texture gradients described by local neighborhoods—a cue identified in human vision literature—are also representable within convolutional neural networks. To this end, we trained both unsupervised and supervised CNN models on the renderings of slanted surfaces with random Polka dot patterns and analyzed their internal latent representations. The results show that the unsupervised models have similar prediction biases as humans across all experiments, while supervised CNN models do not exhibit similar biases. The latent spaces of the unsupervised models can be linearly separated into axes representing field of view and optical slant. For supervised models, this ability varies substantially with model architecture and the kind of supervision (continuous slant vs. sign of slant). Even though this study says nothing of any shared mechanism, these findings suggest that unsupervised CNN models can share similar predictions to the human visual system. Code: github.com/brownvc/Slant-CNN-Biases.

High-quality large-scale scene rendering requires a scalable representation and accurate camera poses. This research combines tile-based hybrid neural fields with parallel distributive optimization to improve bundle-adjusting neural radiance fields. The proposed method scales with a divide-and-conquer strategy. We partition scenes into tiles, each with a multi-resolution hash feature grid and shallow chained diffuse and specular multilayer perceptrons (MLPs). Tiles unify foreground and background via a spatial contraction function that allows both distant objects in outdoor scenes and planar reflections as virtual images outside the tile. Decomposing appearance with the specular MLP allows a specular-aware warping loss to provide a second optimization path for camera poses. We apply the alternating direction method of multipliers (ADMM) to achieve consensus among camera poses while maintaining parallel tile optimization. Experimental results show that our method outperforms state-of-the-art neural scene rendering method quality by 5%--10% in PSNR, maintaining sharp distant objects and view-dependent reflections across six indoor and outdoor scenes.

Immersive visualization in virtual reality (VR) allows us to exploit visual cues for perception in 3D space, yet few existing studies have measured the effects of visual cues. Across a desktop monitor and a head-mounted display (HMD), we assessed scatterplot designs which vary their use of visual cues-motion, shading, perspective (graphical projection), and dimensionality-on two sets of data. We conducted a user study with a summary task in which 32 participants estimated the classification accuracy of an artificial neural network from the scatterplots. With Bayesian multilevel modeling, we capture the intricate visual effects and find that no cue alone explains all the variance in estimation error. Visual motion cues generally reduce participants' estimation error; besides this motion, using other cues may increase participants' estimation error. Using an HMD, adding visual motion cues, providing a third data dimension, or showing a more complicated dataset leads to longer response times. We speculate that most visual cues may not strongly affect perception in immersive analytics unless they change people's mental model about data. In summary, by studying participants as they interpret the output from a complicated machine learning model, we advance our understanding of how to use the visual cues in immersive analytics.

Sharing annotations encourages feedback, discussion, and knowledge passing among readers and can be beneficial for personal and public use. Prior augmented reality (AR) systems have expanded these benefits to both digital and printed documents. However, despite smartphone AR now being widely available, there is a lack of research about how to use AR effectively for interactive document annotation. We propose Dually Noted, a smartphone-based AR annotation system that recognizes the layout of structural elements in a printed document for real-time authoring and viewing of annotations. We conducted experience prototyping with eight users to elicit potential benefits and challenges within smartphone AR, and this informed the resulting Dually Noted system and annotation interactions with the document elements. AR annotation is often unwieldy, but during a 12-user empirical study our novel structural understanding component allows Dually Noted to improve precise highlighting and annotation interaction accuracy by 13%, increase interaction speed by 42%, and significantly lower cognitive load over a baseline method without document layout understanding. Qualitatively, participants commented that Dually Noted was a swift and portable annotation experience. Overall, our research provides new methods and insights for how to improve AR annotations for physical documents.

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

Reconstructing 3D objects in natural environments requires solving the ill-posed problem of geometry, spatially-varying material, and lighting estimation. As such, many approaches impractically constrain to a dark environment, use controlled lighting rigs, or use few handheld captures but suffer reduced quality. We develop a method that uses just two smartphone exposures captured in ambient lighting to reconstruct appearance more accurately and practically than baseline methods. Our insight is that we can use a flash/no-flash RGB-D pair to pose an inverse rendering problem using point lighting. This allows efficient differentiable rendering to optimize depth and normals from a good initialization and so also the simultaneous optimization of diffuse environment illumination and SVBRDF material. We find that this reduces diffuse albedo error by 25%, specular error by 46%, and normal error by 30% against single-and paired-image baselines that use learning-based techniques. Given that our approach is practical for everyday solid objects, we enable photorealistic relighting for mobile photography and easier content creation for augmented reality.

Abstract Recent advances in machine learning have led to increased interest in solving visual computing problems using methods that employ coordinate‐based neural networks. These methods, which we call neural fields , parameterize physical properties of scenes or objects across space and time. They have seen widespread success in problems such as 3D shape and image synthesis, animation of human bodies, 3D reconstruction, and pose estimation. Rapid progress has led to numerous papers, but a consolidation of the discovered knowledge has not yet emerged. We provide context, mathematical grounding, and a review of over 250 papers in the literature on neural fields. In Part I , we focus on neural field techniques by identifying common components of neural field methods, including different conditioning, representation, forward map, architecture, and manipulation methods. In Part II , we focus on applications of neural fields to different problems in visual computing, and beyond (e.g., robotics, audio). Our review shows the breadth of topics already covered in visual computing, both historically and in current incarnations, and highlights the improved quality, flexibility, and capability brought by neural field methods. Finally, we present a companion website that acts as a living database that can be continually updated by the community.

We propose a scalable neural scene reconstruction and rendering method to support distributed training and interactive rendering of large indoor scenes. Our representation is based on tiles. Tile appearances are trained in parallel through a background sampling strategy that augments each tile with distant scene information via a proxy global mesh. Each tile has two low-capacity MLPs: one for view-independent appearance (diffuse color and shading) and one for view-dependent appearance (specular highlights, reflections). We leverage the phenomena that complex view-dependent scene reflections can be attributed to virtual lights underneath surfaces at the total ray distance to the source. This lets us handle sparse samplings of the input scene where reflection highlights do not always appear consistently in input images. We show interactive free-viewpoint rendering results from five scenes, one of which covers an area of more than 100 m 2 . Experimental results show that our method produces higher-quality renderings than a single large-capacity MLP and five recent neural proxy-geometry and voxel-based baseline methods. Our code and data are available at project webpage https://xchaowu.github.io/papers/scalable-nisr.

The dominant markup language for Web visualizations-Scalable Vector Graphics (SVG)-is comparatively easy to learn, and is open, accessible, customizable via CSS, and searchable via the DOM, with easy interaction handling and debugging. Because these attributes allow visualization creators to focus on design on implementation details, tools built on top of SVG, such as D3.js, are essential to the visualization community. However, slow SVG rendering can limit designs by effectively capping the number of on-screen data points, and this can force visualization creators to switch to Canvas or WebGL. These are less flexible (e.g., no search or styling via CSS), and harder to learn. We introduce Scalable Scalable Vector Graphics (SSVG) to reduce these limitations and allow complex and smooth visualizations to be created with SVG. SSVG automatically translates interactive SVG visualizations into a dynamic virtual DOM (VDOM) to bypass the browser's slow 'to specification' rendering by intercepting JavaScript function calls. De-coupling the SVG visualization specification from SVG rendering, and obtaining a dynamic VDOM, creates flexibility and opportunity for visualization system research. SSVG uses this flexibility to free up the main thread for more interactivity and renders the visualization with Canvas or WebGL on a web worker. Together, these concepts create a drop-in JavaScript library which can improve rendering performance by 3-9× with only one line of code added. To demonstrate applicability, we describe the use of SSVG on multiple example visualizations including published visualization research. A free copy of this article, collected data, and source code are available as open science at osf.io/ge8wp.

Given an imperfect predictor, we exploit additional features at test time to improve the predictions made, without retraining and without knowledge of the prediction function. This scenario arises if training labels or data are proprietary, restricted, or no longer available, or if training itself is prohibitively expensive. We assume that the additional features are useful if they exhibit strong statistical dependence to the underlying perfect predictor. Then, we empirically estimate and strengthen the statistical dependence between the initial noisy predictor and the additional features via manifold de-noising. As an example, we show that this approach leads to improvement in real-world visual attribute ranking.

Neural networks can represent and accurately reconstruct radiance fields for static 3D scenes (e.g., NeRF). Several works extend these to dynamic scenes captured with monocular video, with promising performance. However, the monocular setting is known to be an under-constrained problem, and so methods rely on data-driven priors for reconstructing dynamic content. We replace these priors with measurements from a time-of-flight (ToF) camera, and introduce a neural representation based on an image formation model for continuous-wave ToF cameras. Instead of working with processed depth maps, we model the raw ToF sensor measurements to improve reconstruction quality and avoid issues with low reflectance regions, multi-path interference, and a sensor's limited unambiguous depth range. We show that this approach improves robustness of dynamic scene reconstruction to erroneous calibration and large motions, and discuss the benefits and limitations of integrating RGB+ToF sensors now available on modern smartphones.

Tools and interfaces are increasingly expected to be synchronous and distributed to accommodate remote collaboration. Yet, adoption of these techniques for data visualization is low partly because development is difficult: existing collaboration software systems either do not support simultaneous interaction or require expensive redevelopment of existing visualizations. We contribute VisConnect: a web-based synchronous distributed collaborative visualization system that supports most web-based SVG data visualizations, balances system safety with responsiveness, and supports simultaneous interaction from many collaborators. VisConnect works with existing visualization implementations with little-to-no code changes by synchronizing low-level JavaScript events across clients such that visualization updates proceed transparently across clients. This is accomplished via a peer-to-peer system that establishes consensus among clients on the per-element sequence of events, and uses a lock service to grant access over elements to clients. We contribute collaborative extensions of traditional visualization interaction techniques, such as drag, brush, and lasso, and discuss different strategies for collaborative visualization interactions. To demonstrate the utility of VisConnect, we present novel examples of collaborative visualizations in the healthcare domain, remote collaboration with annotation, and show in an education case study for e-learning with 22 participants that students found the ability to remotely collaborate on class activities helpful and enjoyable for understanding concepts. A free copy of this paper and source code are available on OSF at osf.io/ut7e6 and at visconnect.us.

Reconstructing object geometry and material from multiple views typically requires optimization. Differentiable path tracing is an appealing framework as it can reproduce complex appearance effects. However, it is difficult to use due to high computational cost. In this paper, we explore how to use differentiable ray tracing to refine an initial coarse mesh and per-mesh-facet material representation. In simulation, we find that it is possible to reconstruct fine geometric and material detail from low resolution input views, allowing high-quality reconstructions in a few hours despite the expense of path tracing. The reconstructions successfully disambiguate shading, shadow, and global illumination effects such as diffuse interreflection from material properties. We demonstrate the impact of different geometry initializations, including space carving, multi-view stereo, and 3D neural networks. Finally, with input captured using smartphone video and a consumer 360° camera for lighting estimation, we also show how to refine initial reconstructions of real-world objects in unconstrained environments.

Pan and zoom timelines and sliders help us navigate large time series data. However, designing efficient interactions can be difficult. We study pan and zoom methods via crowd-sourced experiments on mobile and computer devices, asking which designs and interactions provide faster target acquisition. We find that visual context should be limited for low-distance navigation, but added for far-distance navigation; that timelines should be oriented along the longer axis, especially on mobile; and that, as compared to default techniques, double click, hold, and rub zoom appear to scale worse with task difficulty, whereas brush and especially ortho zoom seem to scale better. Software and data used in this research are available as open source.

Many 4D light field processing applications rely on superpixel segmentations, for which occlusion-aware view consistency is important. Yet, existing methods often enforce consistency by propagating clusters from a central view only, which can lead to inconsistent superpixels for non-central views. Our proposed approach combines an occlusion-aware angular segmentation in horizontal and vertical EPI spaces with an occlusion-aware clustering and propagation step across all views. Qualitative video demonstrations show that this helps to remove flickering and inconsistent boundary shapes versus the state-of-the-art approach, and quantitative metrics reflect these findings with improved boundary accuracy and view consistency scores.

Smartphone augmented reality (AR) lets users interact with physical and virtual spaces simultaneously. With 3D hand tracking, smartphones become apparatus to grab and move virtual objects directly. Based on design considerations for interaction, mobility, and object appearance and physics, we implemented a prototype for portable 3D hand tracking using a smartphone, a Leap Motion controller, and a computation unit. Following an experience prototyping procedure, 12 researchers used the prototype to help explore usability issues and define the design space. We identified issues in perception (moving to the object, reaching for the object), manipulation (successfully grabbing and orienting the object), and behavioral understanding (knowing how to use the smartphone as a viewport). To overcome these issues, we designed object-based feedback and accommodation mechanisms and studied their perceptual and behavioral effects via two tasks: picking up distant objects, and assembling a virtual house from blocks. Our mechanisms enabled significantly faster and more successful user interaction than the initial prototype in picking up and manipulating stationary and moving objects, with a lower cognitive load and greater user preference. The resulting system---Portal-ble---improves user intuition and aids free-hand interactions in mobile situations.

We propose a new approach for creating omnidirectional mixed reality (OmniMR) from moving-camera 360° video. To insert virtual computer-generated elements into a moving-camera 360° video, we reconstruct camera motion and sparse scene content via structure from motion on stitched equirectangular video (the default output format of current 360° cameras). Then, to plausibly reproduce realworld lighting conditions for these inserted elements, we employ inverse tone mapping to recover high dynamic range environment maps which vary spatially along the camera path. We implement our approach into the Unity rendering engine for real-time object rendering with dynamic lighting and user interaction. This expands the use and flexibility of 360° video for mixed reality.

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.

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.

We examine the relationship between eye gaze and typing, focusing on the differences between touch and non-touch typists. To enable typing-based research, we created a 51-participant benchmark dataset for user input across multiple tasks, including user input data, screen recordings, webcam video of the participant's face, and eye tracking positions. There are patterns of eye movements that differ between the two types of typists, representing glances at the keyboard, which can be used to identify touch-.typed strokes with 92% accuracy. Then, we relate eye gaze with cursor activity, aligning both pointing and typing to eye gaze. One demonstrative application of the work is in extending WebGazer, a real-time web-browser-based webcam eye tracker. We show that incorporating typing behavior as a secondary signal improves eye tracking accuracy by 16% for touch typists, and 8% for non-touch typists.

Current unsupervised image-to-image translation techniques struggle to focus their attention on individual objects without altering the background or the way multiple objects interact within a scene. Motivated by the important role of attention in human perception, we tackle this limitation by introducing unsupervised attention mechanisms which are jointly adversarially trained with the generators and discriminators. We empirically demonstrate that our approach is able to attend to relevant regions in the image without requiring any additional supervision, and that by doing so it achieves more realistic mappings compared to recent approaches.

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 for test-time combination of a set of reference predictors with unknown parametric forms. Existing multi-task and transfer learning algorithms focus on training-time transfer and combination, where the parametric forms of predictors are known and shared. However, when the parametric form of a predictor is unknown, e.g., for a human predictor or a predictor in a precompiled library, existing algorithms are not applicable. Instead, we empirically evaluate predictors on sampled data points to measure distances between different predictors. This embeds the set of reference predictors into a Riemannian manifold, upon which we perform manifold denoising to obtain the refined predictor. This allows our approach to make no assumptions about the underlying predictor forms. Our test-time combination algorithm equals or outperforms existing multi-task and transfer learning algorithms on challenging real-world datasets, without introducing specific model assumptions.

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.

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.

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.

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.

Improving the quality of degraded images is a key problem in image processing, but the breadth of the problem leads to domain-specific approaches for tasks such as super-resolution and compression artifact removal. Recent approaches have shown that a general approach is possible by learning application-specific models from examples; however, learning models sophisticated enough to generate high-quality images is computationally expensive, and so specific per-application or per-dataset models are impractical. To solve this problem, we present an efficient semi-local approximation scheme to large-scale Gaussian processes. This allows efficient learning of task-specific image enhancements from example images without reducing quality. As such, our algorithm can be easily customized to specific applications and datasets, and we show the efficiency and effectiveness of our approach across five domains: single-image super-resolution for scene, human face, and text images, and artifact removal in JPEG- and JPEG 2000-encoded images.

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.

Abstract It is now possible to capture the 3D motion of the human body on consumer hardware and to puppet in real time skeleton‐based virtual characters. However, many characters do not have humanoid skeletons. Characters such as spiders and caterpillars do not have boned skeletons at all, and these characters have very different shapes and motions. In general, character control under arbitrary shape and motion transformations is unsolved ‐ how might these motions be mapped? We control characters with a method which avoids the rigging‐skinning pipeline — source and target characters do not have skeletons or rigs. We use interactively‐defined sparse pose correspondences to learn a mapping between arbitrary 3D point source sequences and mesh target sequences. Then, we puppet the target character in real time. We demonstrate the versatility of our method through results on diverse virtual characters with different input motion controllers. Our method provides a fast, flexible, and intuitive interface for arbitrary motion mapping which provides new ways to control characters for real‐time animation.

One fundamental assumption in object recognition as well as in other computer vision and pattern recognition problems is that the data generation process lies on a manifold and that it respects the intrinsic geometry of the manifold. This assumption is held in several successful algorithms for diffusion and regularization, in particular, in graph-Laplacian-based algorithms. We claim that the performance of existing algorithms can be improved if we additionally account for how the manifold is embedded within the ambient space, i.e., if we consider the extrinsic geometry of the manifold. We present a procedure for characterizing the extrinsic (as well as intrinsic) curvature of a manifold M which is described by a sampled point cloud in a high-dimensional Euclidean space. Once estimated, we use this characterization in general diffusion and regularization on M, and form a new regularizer on a point cloud. The resulting re-weighted graph Laplacian demonstrates superior performance over classical graph Laplacian in semi-supervised learning and spectral clustering.

Video collections of places show contrasts and changes in our world, but current interfaces to video collections make it hard for users to explore these changes. Recent state-of-the-art interfaces attempt to solve this problem for 'outside->in' collections, but cannot connect 'inside->out' collections of the same place which do not visually overlap. We extend the focus+context paradigm to create a video-collections+context interface by embedding videos into a panorama. We build a spatio-temporal index and tools for fast exploration of the space and time of the video collection. We demonstrate the flexibility of our representation with interfaces for desktop and mobile flat displays, and for a spherical display with joypad and tablet controllers. We study with users the effect of our video-collection+context system to spatio-temporal localization tasks, and find significant improvements to accuracy and completion time in visual search tasks compared to existing systems. We measure the usability of our interface with System Usability Scale (SUS) and task-specific questionnaires, and find our system scores higher.

Emerging interfaces for video collections of places attempt to link similar content with seamless transitions. However, the automatic computer vision techniques that enable these transitions have many failure cases which lead to artifacts in the final rendered transition. Under these conditions, which transitions are preferred by participants and which artifacts are most objectionable? We perform an experiment with participants comparing seven transition types, from movie cuts and dissolves to image-based warps and virtual camera transitions, across five scenes in a city. This document describes how we condition this experiment on slight and considerable view change cases, and how we analyze the feedback from participants to find their preference for transition types and artifacts. We discover that transition preference varies with view change, that automatic rendered transitions are significantly preferred even with some artifacts, and that dissolve transitions are comparable to less-sophisticated rendered transitions. This leads to insights into what visual features are important to maintain in a rendered transition, and to an artifact ordering within our transitions.

High dynamic range reconstruction of dynamic scenes requires careful handling of dynamic objects to prevent ghosting. However, in a recent review, Srikantha et al. [2012] conclude that "there is no single best method and the selection of an approach depends on the user's goal". We attempt to solve this problem with a novel approach that models the noise distribution of color values. We estimate the likelihood that a pair of colors in different images are observations of the same irradiance, and we use a Markov random field prior to reconstruct irradiance from pixels that are likely to correspond to the same static scene object. Dynamic content is handled by selecting a single low dynamic range source image and hand-held capture is supported through homography-based image alignment. Our noise-based reconstruction method achieves better ghost detection and removal than state-of-the-art methods for cluttered scenes with large object displacements. As such, our method is broadly applicable and helps move the field towards a single method for dynamic scene HDR reconstruction.

Lenticular prints are a popular medium for producing automultiscopic glasses-free 3D images. The light field emitted by such prints has a fixed spatial and angular resolution. We increase both perceived angular and spatial resolution by modifying the lenslet array to better match the content of a given light field. Our optimization algorithm analyzes the input light field and computes an optimal lenslet size, shape, and arrangement that best matches the input light field given a set of output parameters. The resulting emitted light field shows higher detail and smoother motion parallax compared to fixed-size lens arrays. We demonstrate our technique using rendered simulations and by 3D printing lens arrays, and we validate our approach in simulation with a user study.

Abstract Removing dynamic objects from videos is an extremely challenging problem that even visual effects professionals often solve with time‐consuming manual frame‐by‐frame editing. We propose a new approach to video completion that can deal with complex scenes containing dynamic background and non‐periodical moving objects. We build upon the idea that the spatio‐temporal hole left by a removed object can be filled with data available on other regions of the video where the occluded objects were visible. Video completion is performed by solving a large combinatorial problem that searches for an optimal pattern of pixel offsets from occluded to unoccluded regions. Our contribution includes an energy functional that generalizes well over different scenes with stable parameters, and that has the desirable convergence properties for a graph‐cut‐based optimization. We provide an interface to guide the completion process that both reduces computation time and allows for efficient correction of small errors in the result. We demonstrate that our approach can effectively complete complex, high‐resolution occlusions that are greater in difficulty than what existing methods have shown.

Abstract Photographs usually show a scene from a single perspective. However, as commonly seen in art, scenes and objects can be visualized from multiple perspectives. Making such images manually is time consuming and tedious. We propose a novel system for designing multi‐perspective images and videos. First, the images in the input sequence are aligned using structure from motion. This enables us to track feature points across the sequence. Second, the user chooses portal polygons in a target image into which different perspectives are to be embedded. The corresponding image regions from the other images are then copied into these portals. Due to the tracking feature and automatic warping, this approach is considerably faster than current tools. We explore a wide range of artistic applications using our system with image and video data, such as looking around corners and up and down stair cases, recursive multi‐perspective imaging, cubism and panoramas.

The abundance of mobile devices and digital cameras with video capture makes it easy to obtain large collections of video clips that contain the same location, environment, or event. However, such an unstructured collection is difficult to comprehend and explore. We propose a system that analyzes collections of unstructured but related video data to create a Videoscape : a data structure that enables interactive exploration of video collections by visually navigating -- spatially and/or temporally -- between different clips. We automatically identify transition opportunities, or portals . From these portals, we construct the Videoscape, a graph whose edges are video clips and whose nodes are portals between clips. Now structured, the videos can be interactively explored by walking the graph or by geographic map. Given this system, we gauge preference for different video transition styles in a user study, and generate heuristics that automatically choose an appropriate transition style. We evaluate our system using three further user studies, which allows us to conclude that Videoscapes provides significant benefits over related methods. Our system leads to previously unseen ways of interactive spatio-temporal exploration of casually captured videos, and we demonstrate this on several video collections.

We present a method to synthesize plausible video sequences of humans according to user-defined body motions and viewpoints. We first capture a small database of multi-view video sequences of an actor performing various basic motions. This database needs to be captured only once and serves as the input to our synthesis algorithm. We then apply a marker-less model-based performance capture approach to the entire database to obtain pose and geometry of the actor in each database frame. To create novel video sequences of the actor from the database, a user animates a 3D human skeleton with novel motion and viewpoints. Our technique then synthesizes a realistic video sequence of the actor performing the specified motion based only on the initial database. The first key component of our approach is a new efficient retrieval strategy to find appropriate spatio-temporally coherent database frames from which to synthesize target video frames. The second key component is a warping-based texture synthesis approach that uses the retrieved most-similar database frames to synthesize spatio-temporally coherent target video frames. For instance, this enables us to easily create video sequences of actors performing dangerous stunts without them being placed in harm's way. We show through a variety of result videos and a user study that we can synthesize realistic videos of people, even if the target motions and camera views are different from the database content.

We present a novel integration of a brain-computer interface (BCI) with a multi-touch surface. BCIs based on the P300 paradigm often use a visual stimulus of a flashing character to elicit an event related potential in the brain's EEG signal. Traditionally, P300-based BCI paradigms use a grid layout of visual targets, commonly an alphabet, and allow users to select targets using their thoughts. In our new system a multi-touch table senses objects placed upon its surface and the system can highlight the objects on the table by flashing an area of light around them. This allows us to construct a P300-based BCI that uses a user-assembled collection of objects as targets, rather than a pre-determined grid layout. An experiment shows that our new paradigm works just as well as the traditional paradigms, thus highlighting the potential for BCIs to be integrated in a broader range of situations.