Anamorphosis refers to a category of images that are intentionally distorted, making them unrecognizable when viewed directly. Their true form only reveals itself when seen from a specific viewpoint, which can be through some catadioptric device like a mirror or a lens. While the construction of these mathematical devices can be traced back to as early as the 17th century [28], they are only interpretable when viewed from a specific vantage point and tend to lose meaning when seen normally. In this paper, we revisit these famous optical illusions with a generative twist. With the help of latent rectified flow models, we propose a method to create anamorphic images that still retain a valid interpretation when viewed directly. To this end, we introduce Laplacian Pyramid Warping, a frequency-aware image warping technique key to generating high-quality visuals. Our work extends Visual Anagrams [17] to latent space models and to a wider range of spatial transforms, enabling the creation of novel generative perceptual illusions.

Time series data is ubiquitous in people’s daily production and life, and visualizations augmented with annotations can significantly facilitate the understanding of such data and promote downstream tasks. Consequently, numerous annotation tools have been developed to detect and narrate useful patterns within time series visualizations. However, most existing tools can only identify basic factual insights (e.g., increasing or decreasing trends) that are already present in the charts. When users need deeper insights (e.g., predicting future trends) and richer contextual information (e.g., associative patterns between dimensions within and beyond the chart), these tools often fall short. To address this challenge, we present AutoMA, a system that automatically generates multi-level annotations for time series visualizations. We introduce an LLM-based pattern extraction method that supports the identification of seven distinct temporal patterns. Furthermore, we present a multi-level annotation design space that encompasses six specific annotation tasks, aimed at delivering richer contextual information. The generated annotations span a spectrum of information, ranging from directly observable temporal patterns to deeper insights obtained through further computation, and ultimately to advanced inter-dimensional association patterns. Finally, we demonstrate the effectiveness of our approach through experiments and user evaluations. The results indicate that AutoMA significantly enhances users’ ability to comprehend and explore time series data.

Humans encounter a vast array of sensory stimuli in their everyday lives. However, many visualization techniques primarily utilize visual feedback, which may disregard certain intricate details. Relying on a single visual channel may overlook complex layouts. However, how haptic force feedback can be used to assist visualization remained under-explored. In this work, we initially conducted a literature review to identify potential problems in the visualization of large datasets and engaged in discussions with domain experts to explore the potential of haptic force feedback and visual collision representation. Subsequently, we designed an innovative haptic force feedback knob, which included 3 primary modules and 29 elements. To evaluate the clarity and usefulness of this design space, we conducted a workshop and devised “recommended solutions” for the identified visualization problems. Finally, we implemented a prototype of the haptic force feedback knob and assessed its performance on scatterplot and parallel coordinate plot tasks using large datasets. The results indicated that the knob prototype could reduce visual strain and enhance the efficiency of visualization tasks.

Visual analysis has been increasingly integrated into the exploration of temporal networks, as visualization methods have the capability to present time-varying attributes and relationships of entities in an easy-to-read manner. Visualization techniques have been employed in a variety of dynamic network datasets, including social media networks, academic citation networks, and financial transaction networks. However, effectively visualizing dynamic basketball player network data, which consists of numerical networks, intensive timestamps, and subtle changes, remains a challenge for analysts. To address this issue, we propose a snapshot extraction algorithm that involves human-in-the-loop methodology to help users divide a series of networks into hierarchical snapshots for subsequent network analysis tasks, such as node exploration and network pattern analysis. Furthermore, we design and implement a prototype system, called DBNetVizor, for dynamic basketball player network data visualization. DBNetVizor integrates a graphical user interface to help users extract snapshots visually and interactively, as well as multiple linked visualization charts to display macro- and micro-level information of dynamic basketball player network data. To demonstrate the usability and efficiency of our proposed methods, we present two case studies based on dynamic basketball player network data in a competition. Additionally, we conduct an evaluation and receive positive feedback.

In the complex landscape of the Internet, inter-domain routing systems are essential for ensuring seamless connectivity and reachability across autonomous systems. However, the lack of dependable security validation mechanisms in these systems poses persistent challenges. Vulnerabilities such as prefix hijacking, path forgery, and route leakage not only compromise network operators and users, but also threaten the stability and accessibility of the Internet’s core infrastructure. To address this, visualization and visual analytics techniques are adept at identifying and detecting security threats, offering network administrators effective methods to monitor and maintain network operations. This paper presents a comprehensive survey of the state-of-the-art research in visualization and visual analytics for inter-domain routing security. We delineate four scenarios for tasks analysis in network visualization: monitoring, detection, verification, and discovery. Each category is explored in detail, focusing on the employed data sources and visualization techniques. Several key findings are presented at the end of each category, aimed at providing researchers and practitioners with research inspiration. Furthermore, we examine the trends of academic interest observed in recent decades and propose potential directions for future research in visual analytics pertaining to Internet infrastructure security.

The intricate geometric complexity of knots, tangles, dreads and clumps require sophisticated grooming systems that allow artists to both realistically model and artistically control fur and hair systems. Recent volumetric and 3D neural style transfer techniques provided a new paradigm of art directability, allowing artists to modify assets drastically with the use of single style images. However, these previous 3D neural stylization approaches were limited to volumes and meshes. In this paper we propose the first stylization pipeline to support hair and fur. Through a carefully tailored fur/hair representation, our approach allows complex, 3D consistent and temporally coherent grooms that are stylized using style images.

Identifying causality behind complex systems plays a significant role in different domains, such as decision-making, policy implementations, and management recommendations. However, existing causality studies on temporal event sequence data mainly focus on individual causal discovery, which is incapable of capturing combined causality. To address the gap in combined causality discovery on temporal event sequence data, eliminating and recruiting principles are defined to balance the effectiveness and controllability of cause combinations. We also leverage the Granger causality algorithm based on the Reactive point processes to describe impelling or inhibiting behavior patterns among entities. In addition, we design an informative and aesthetic visual metaphor of "electrocircuit" to encode aggregated causality for ensuring that our causality visualization exhibits no node-overlap, no edge-intersection, and no link-ambiguity. Aggregation layout, diverse sorting strategies, and smooth interactions are also integrated into our directed, weighted, and parallel-based hypergraph for illustrating combined causality. Our developed combined causality visual analysis system, namely VAC$^{2}$2, can help users effectively explore combined causes as well as individual causes. This interactive system supports multi-level causality exploration with diverse ordering strategies and a focus+context technique to help users obtain different levels of information abstraction. The usefulness and effectiveness of our work are further evaluated by conducting two case studies and a controlled user study on event sequence data.

Video editing and generation methods often rely on pre-trained image-based diffusion models. During the diffusion process, however, the reliance on rudimentary noise sampling techniques that do not preserve correlations present in subsequent frames of a video is detrimental to the quality of the results. This either produces high-frequency flickering, or texture-sticking artifacts that are not amenable to post-processing. With this in mind, we propose a novel method for preserving temporal correlations in a sequence of noise samples. This approach is materialized by a novel noise representation, dubbed $\int$-noise (integral noise), that reinterprets individual noise samples as a continuously integrated noise field: pixel values do not represent discrete values, but are rather the integral of an underlying infinite-resolution noise over the pixel area. Additionally, we propose a carefully tailored transport method that uses $\int$-noise to accurately advect noise samples over a sequence of frames, maximizing the correlation between different frames while also preserving the noise properties. Our results demonstrate that the proposed $\int$-noise can be used for a variety of tasks, such as video restoration, surrogate rendering, and conditional video generation.

Abstract An ongoing challenge in fluid animation is the faithful preservation of vortical details, which impacts the visual depiction of flows. We propose the Impulse Particle‐In‐Cell (IPIC) method, a novel extension of the popular Affine Particle‐In‐Cell (APIC) method that makes use of the impulse gauge formulation of the fluid equations. Our approach performs a coupled advection‐stretching during particle‐based advection to better preserve circulation and vortical details. The associated algorithmic changes are simple and straightforward to implement, and our results demonstrate that the proposed method is able to achieve more energetic and visually appealing smoke and liquid flows than APIC.

Video data refers to digital information in the form of a series of frames or images representing continuous motion captured by a video recording device. In various domains such as security, sports, education, and entertainment, a significant amount of video data is generated and stored daily. However, analyzing these videos manually is challenging due to their intrinsic characteristics, including large-scale, redundancy, contextual dependencies, and multimodality. Consequently, researchers have extensively explored visualization techniques to address these complexities. In this investigation, we review the state-of-the-art techniques in video visualization and visual analysis. Initially, we provide an overview of the design space for video visualization and visual analysis techniques. Subsequently, we organize and classify these techniques based on visual analysis tasks and application scenarios, providing detailed descriptions within each category. Drawing upon a comprehensive review of existing research, we provide a critical evaluation and propose potential opportunities for future research. Additionally, we have developed a web-based survey browser for convenient exploration of our created classification framework and the associated scholarly articles (https://zjutvis.github.io/VOVideo/).

Abstract Controlling fluid simulations is notoriously difficult due to its high computational cost and the fact that user control inputs can cause unphysical motion. We present an interactive method for deformation‐based fluid control. Our method aims at balancing the direct deformations of fluid fields and the preservation of physical characteristics. We train convolutional neural networks with physics‐inspired loss functions together with a differentiable fluid simulator, and provide an efficient workflow for flow manipulations at test time. We demonstrate diverse test cases to analyze our carefully designed objectives and show that they lead to physical and eventually visually appealing modifications on edited fluid data.

Machine vision measurement (MVM) is an essential approach that measures the area or length of a target efficiently and non-destructively for product quality control. The result of MVM is determined by its configuration, especially the lighting scheme design in image acquisition and the algorithmic parameter optimization in image processing. In a traditional workflow, engineers constantly adjust and verify the configuration for an acceptable result, which is time-consuming and significantly depends on expertise. To address these challenges, we propose a target-independent approach, visual interactive image clustering, which facilitates configuration optimization by grouping images into different clusters to suggest lighting schemes with common parameters. Our approach has four steps: data preparation, data sampling, data processing, and visual analysis with our visualization system. During preparation, engineers design several candidate lighting schemes to acquire images and develop an algorithm to process images. Our approach samples engineer-defined parameters for each image and obtains results by executing the algorithm. The core of data processing is the explainable measurement of the relationships among images using the algorithmic parameters. Based on the image relationships, we develop VMExplorer, a visual analytics system that assists engineers in grouping images into clusters and exploring parameters. Finally, engineers can determine an appropriate lighting scheme with robust parameter combinations. To demonstrate the effectiveness and usability of our approach, we conduct a case study with engineers and obtain feedback from expert interviews.

Solving linear system of equations stemming from Laplacian operators is at the heart of a wide range of applications. Due to the sparsity of the linear systems, iterative solvers such as Conjugate Gradient and Multigrid are usually employed when the solution has a large number of degrees of freedom. These iterative solvers can be seen as sparse approximations of the Green’s function for the Laplacian operator. In this paper we propose a machine learning approach that regresses a Green’s function from boundary conditions. This is enabled by a Green’s function that can be effectively represented in a multi-scale fashion, drastically reducing the cost associated with a dense matrix representation. Additionally, since the Green’s function is solely dependent on boundary conditions, training the proposed neural network does not require sampling the right-hand side of the linear system. We show results that our method outperforms state of the art Conjugate Gradient and Multigrid methods.

Abstract Creative processes of artists often start with hand‐drawn sketches illustrating an object. Pre‐visualizing these keyframes is especially challenging when applied to volumetric materials such as smoke. The authored 3D density volumes must capture realistic flow details and turbulent structures, which is highly non‐trivial and remains a manual and time‐consuming process. We therefore present a method to compute a 3D smoke density field directly from 2D artist sketches, bridging the gap between early‐stage prototyping of smoke keyframes and pre‐visualization. From the sketch inputs, we compute an initial volume estimate and optimize the density iteratively with an updater CNN. Our differentiable sketcher is embedded into the end‐to‐end training, which results in robust reconstructions. Our training data set and sketch augmentation strategy are designed such that it enables general applicability. We evaluate the method on synthetic inputs and sketches from artists depicting both realistic smoke volumes and highly non‐physical smoke shapes. The high computational performance and robustness of our method at test time allows interactive authoring sessions of volumetric density fields for rapid prototyping of ideas by novice users.

Abstract Fluid control often uses optimization of control forces that are added to a simulation at each time step, such that the final animation matches a single or multiple target density keyframes provided by an artist. The optimization problem is strongly under‐constrained with a high‐dimensional parameter space, and finding optimal solutions is challenging, especially for higher resolution simulations. In this paper, we propose two novel ideas that jointly tackle the lack of constraints and high dimensionality of the parameter space. We first consider the fact that optimized forces are allowed to have divergent modes during the optimization process. These divergent modes are not entirely projected out by the pressure solver step, manifesting as unphysical smoke sources that are explored by the optimizer to match a desired target. Thus, we reduce the space of the possible forces to the family of strictly divergence‐free velocity fields, by optimizing directly for a vector potential. We synergistically combine this with a smoothness regularization based on a spectral decomposition of control force fields. Our method enforces lower frequencies of the force fields to be optimized first by filtering force frequencies in the Fourier domain. The mask‐growing strategy is inspired by Kolmogorov's theory about scales of turbulence. We demonstrate improved results for 2D and 3D fluid control especially in higher‐resolution settings, while eliminating the need for manual parameter tuning. We showcase various applications of our method, where the user effectively creates or edits smoke simulations.

The goal of natural image matting is the estimation of opacities of a user-defined foreground object that is essential in creating realistic composite imagery. Natural matting is a challenging process due to the high number of unknowns in the mathematical modeling of the problem, namely the opacities as well as the foreground and background layer colors, while the original image serves as the single observation. In this paper, we propose the estimation of the layer colors through the use of deep neural networks prior to the opacity estimation. The layer color estimation is a better match for the capabilities of neural networks, and the availability of these colors substantially increase the performance of opacity estimation due to the reduced number of unknowns in the compositing equation. A prominent approach to matting in parallel to ours is called sampling-based matting, which involves gathering color samples from known-opacity regions to predict the layer colors. Our approach outperforms not only the previous hand-crafted sampling algorithms, but also current data-driven methods. We hence classify our method as a hybrid sampling- and learning-based approach to matting, and demonstrate the effectiveness of our approach through detailed ablation studies using alternative network architectures.