Topic overview

Graphics

948 works3367 researchers0 institutions

Computer Vision

Topic snapshot

What this area looks like now

948works

3367authors

0experts visible

0communities

Next steps

Move from topic reading into action

The graph preview below keeps the nearby papers, people and communities visible in the same reading flow.

Inspect nearby papers, researchers, institutions and communities without opening a separate graph page.

Building this graph slice

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2013arXiv

A Dual-Beam Method-of-Images 3D Searchlight BSSRDF

We present a novel BSSRDF for rendering translucent materials. Angular effects lacking in previous BSSRDF models are incorporated by using a dual-beam formulation. We employ a Placzek's Lemma interpretation of the method of images and discard diffusion theory. Instead, we derive a plane-parallel transformation of the BSSRDF to form the associated BRDF and optimize the image confiurations such that the BRDF is close to the known analytic solutions for the associated albedo problem. This ensures reciprocity, accurate colors, and provides an automatic level-of-detail transition for translucent objects that appear at various distances in an image. Despite optimizing the subsurface fluence in a plane-parallel setting, we find that this also leads to fairly accurate fluence distributions throughout the volume in the original 3D searchlight problem. Our method-of-images modifications can also improve the accuracy of previous BSSRDFs.

preprint2018arXiv

A reciprocal formulation of non-exponential radiative transfer. 1: Sketch and motivation

Previous proposals to permit non-exponential free-path statistics in radiative transfer have not included support for volume and boundary sources that are spatially uncorrelated from the scattering events in the medium. Birth-collision free paths are treated identically to collision-collision free paths and application of this to general, bounded scenes with inclusions leads to non-reciprocal transport. Beginning with reciprocity as a desired property, we propose a new way to integrate non-exponential transport theory into general scenes. We distinguish between the free-path-length statistics between correlated medium particles and the free-path-length statistics beginning at locations not correlated to medium particles, such as boundary surfaces, inclusions and uncorrelated sources. Reciprocity requires that the uncorrelated free-path distributions are simply the normalized transmittance of the correlated free-path distributions. The combination leads to an equilibrium imbedding of a previously derived generalized transport equation into bounded domains. We compare predictions of this approach to Monte Carlo simulation of multiple scattering from negatively-correlated suspensions of monodispersive hard spheres in bounded two-dimensional domains and demonstrate improved performance relative to previous work. We also derive new, exact, reciprocal, single-scattering solutions for plane-parallel half-spaces over a variety of non-exponential media types.

preprint2018arXiv

Isotropic Scattering in a Flatland Half-Space

We solve the Milne, constant-source and albedo problems for isotropic scattering in a two-dimensional "Flatland" half-space via the Wiener-Hopf method. The Flatland $H$-function is derived and benchmark values and some identities unique to Flatland are presented. A number of the derivations are supported by Monte Carlo simulation.

preprint2018arXiv

DeepWarp: DNN-based Nonlinear Deformation

DeepWarp is an efficient and highly re-usable deep neural network (DNN) based nonlinear deformable simulation framework. Unlike other deep learning applications such as image recognition, where different inputs have a uniform and consistent format (e.g. an array of all the pixels in an image), the input for deformable simulation is quite variable, high-dimensional, and parametrization-unfriendly. Consequently, even though DNN is known for its rich expressivity of nonlinear functions, directly using DNN to reconstruct the force-displacement relation for general deformable simulation is nearly impossible. DeepWarp obviates this difficulty by partially restoring the force-displacement relation via warping the nodal displacement simulated using a simplistic constitutive model -- the linear elasticity. In other words, DeepWarp yields an incremental displacement fix based on a simplified (therefore incorrect) simulation result other than returning the unknown displacement directly. We contrive a compact yet effective feature vector including geodesic, potential and digression to sort training pairs of per-node linear and nonlinear displacement. DeepWarp is robust under different model shapes and tessellations. With the assistance of deformation substructuring, one DNN training is able to handle a wide range of 3D models of various geometries including most examples shown in the paper. Thanks to the linear elasticity and its constant system matrix, the underlying simulator only needs to perform one pre-factorized matrix solve at each time step, and DeepWarp is able to simulate large models in real time.

preprint2019arXiv

Dialog on a canvas with a machine

We propose a new form of human-machine interaction. It is a pictorial game consisting of interactive rounds of creation between artists and a machine. They repetitively paint one after the other. At its rounds, the computer partially completes the drawing using machine learning algorithms, and projects its additions directly on the canvas, which the artists are free to insert or modify. Alongside fostering creativity, the process is designed to question the growing interaction between humans and machines.

preprint2019arXiv

Statistical Analysis and Modeling of the Geometry and Topology of Plant Roots

The root is an important organ of a plant since it is responsible for water and nutrient uptake. Analyzing and modelling variabilities in the geometry and topology of roots can help in assessing the plant's health, understanding its growth patterns, and modeling relations between plant species and between plants and their environment. In this article, we develop a framework for the statistical analysis and modeling of the geometry and topology of plant roots. We represent root structures as points in a tree-shape space equipped with a metric that quantifies geometric and topological differences between pairs of roots. We then use these building blocks to compute geodesics, i.e., optimal deformations under the metric between root structures, and to perform statistical analysis on root populations. We demonstrate the utility of the proposed framework through an application to a dataset of wheat roots grown in different environmental conditions. We also show that the framework can be used in various applications including classification and regression.

preprint2019arXiv

Inverse Halftoning Through Structure-Aware Deep Convolutional Neural Networks

The primary issue in inverse halftoning is removing noisy dots on flat areas and restoring image structures (e.g., lines, patterns) on textured areas. Hence, a new structure-aware deep convolutional neural network that incorporates two subnetworks is proposed in this paper. One subnetwork is for image structure prediction while the other is for continuous-tone image reconstruction. First, to predict image structures, patch pairs comprising continuous-tone patches and the corresponding halftoned patches generated through digital halftoning are trained. Subsequently, gradient patches are generated by convolving gradient filters with the continuous-tone patches. The subnetwork for the image structure prediction is trained using the mini-batch gradient descent algorithm given the halftoned patches and gradient patches, which are fed into the input and loss layers of the subnetwork, respectively. Next, the predicted map including the image structures is stacked on the top of the input halftoned image through a fusion layer and fed into the image reconstruction subnetwork such that the entire network is trained adaptively to the image structures. The experimental results confirm that the proposed structure-aware network can remove noisy dot-patterns well on flat areas and restore details clearly on textured areas. Furthermore, it is demonstrated that the proposed method surpasses the conventional state-of-the-art methods based on deep convolutional neural networks and locally learned dictionaries.

preprint2020arXiv

A Feature-aware SPH for Isotropic Unstructured Mesh Generation

In this paper, we present a feature-aware SPH method for the concurrent and automated isotropic unstructured mesh generation. Two additional objectives are achieved with the proposed method compared to the original SPH-based mesh generator (Fu et al., 2019). First, a feature boundary correction term is introduced to address the issue of incomplete kernel support at the boundary vicinity. The mesh generation of feature curves, feature surfaces and volumes can be handled concurrently without explicitly following a dimensional sequence. Second, a two-phase model is proposed to characterize the mesh-generation procedure by a feature-size-adaptation phase and a mesh-quality-optimization phase. By proposing a new error measurement criterion and an adaptive control system with two sets of simulation parameters, the objectives of faster feature-size adaptation and local mesh-quality improvement are merged into a consistent framework. The proposed method is validated with a set of 2D and 3D numerical tests with different complexities and scales. The results demonstrate that high-quality meshes are generated with a significant speedup of convergence.

preprint2020arXiv

A Hybrid Lagrangian/Eulerian Collocated Advection and Projection Method for Fluid Simulation

We present a hybrid particle/grid approach for simulating incompressible fluids on collocated velocity grids. We interchangeably use particle and grid representations of transported quantities to balance efficiency and accuracy. A novel Backward Semi-Lagrangian method is derived to improve accuracy of grid based advection. Our approach utilizes the implicit formula associated with solutions of Burgers' equation. We solve this equation using Newton's method enabled by $C^1$ continuous grid interpolation. We enforce incompressibility over collocated, rather than staggered grids. Our projection technique is variational and designed for B-spline interpolation over regular grids where multiquadratic interpolation is used for velocity and multilinear interpolation for pressure. Despite our use of regular grids, we extend the variational technique to allow for cut-cell definition of irregular flow domains for both Dirichlet and free surface boundary conditions.

preprint2020arXiv

A Survey on Deep Learning Techniques for Stereo-based Depth Estimation

Estimating depth from RGB images is a long-standing ill-posed problem, which has been explored for decades by the computer vision, graphics, and machine learning communities. Among the existing techniques, stereo matching remains one of the most widely used in the literature due to its strong connection to the human binocular system. Traditionally, stereo-based depth estimation has been addressed through matching hand-crafted features across multiple images. Despite the extensive amount of research, these traditional techniques still suffer in the presence of highly textured areas, large uniform regions, and occlusions. Motivated by their growing success in solving various 2D and 3D vision problems, deep learning for stereo-based depth estimation has attracted growing interest from the community, with more than 150 papers published in this area between 2014 and 2019. This new generation of methods has demonstrated a significant leap in performance, enabling applications such as autonomous driving and augmented reality. In this article, we provide a comprehensive survey of this new and continuously growing field of research, summarize the most commonly used pipelines, and discuss their benefits and limitations. In retrospect of what has been achieved so far, we also conjecture what the future may hold for deep learning-based stereo for depth estimation research.

preprint2020arXiv

Deep Vectorization of Technical Drawings

We present a new method for vectorization of technical line drawings, such as floor plans, architectural drawings, and 2D CAD images. Our method includes (1) a deep learning-based cleaning stage to eliminate the background and imperfections in the image and fill in missing parts, (2) a transformer-based network to estimate vector primitives, and (3) optimization procedure to obtain the final primitive configurations. We train the networks on synthetic data, renderings of vector line drawings, and manually vectorized scans of line drawings. Our method quantitatively and qualitatively outperforms a number of existing techniques on a collection of representative technical drawings.

preprint2020arXiv

Michelson Holography: Dual-SLM Holography with Camera-in-the-loop Optimization

We introduce Michelson Holography (MH), a holographic display technology that optimizes image quality for emerging holographic near-eye displays. Using two spatial light modulators, MH is capable of leveraging destructive interference to optically cancel out undiffracted light corrupting the observed image. We calibrate this system using emerging camera-in-the-loop holography techniques and demonstrate state-of-the-art holographic 2D image quality.

preprint2020arXiv

Visualization of Human Spine Biomechanics for Spinal Surgery

We propose a visualization application, designed for the exploration of human spine simulation data. Our goal is to support research in biomechanical spine simulation and advance efforts to implement simulation-backed analysis in surgical applications. Biomechanical simulation is a state-of-the-art technique for analyzing load distributions of spinal structures. Through the inclusion of patient-specific data, such simulations may facilitate personalized treatment and customized surgical interventions. Difficulties in spine modelling and simulation can be partly attributed to poor result representation, which may also be a hindrance when introducing such techniques into a clinical environment. Comparisons of measurements across multiple similar anatomical structures and the integration of temporal data make commonly available diagrams and charts insufficient for an intuitive and systematic display of results. Therefore, we facilitate methods such as multiple coordinated views, abstraction and focus and context to display simulation outcomes in a dedicated tool. By linking the result data with patient-specific anatomy, we make relevant parameters tangible for clinicians. Furthermore, we introduce new concepts to show the directions of impact force vectors, which were not accessible before. We integrated our toolset into a spine segmentation and simulation pipeline and evaluated our methods with both surgeons and biomechanical researchers. When comparing our methods against standard representations that are currently in use, we found increases in accuracy and speed in data exploration tasks. In a qualitative review, domain experts deemed the tool highly useful when dealing with simulation result data, which typically combines time-dependent patient movement and the resulting force distributions on spinal structures.

preprint2020arXiv

Diptychs of human and machine perceptions

We propose visual creations that put differences in algorithms and humans \emph{perceptions} into perspective. We exploit saliency maps of neural networks and visual focus of humans to create diptychs that are reinterpretations of an original image according to both machine and human attentions. Using those diptychs as a qualitative evaluation of perception, we discuss some crucial issues of current \textit{task-oriented} artificial intelligence.

preprint2020arXiv

Eyes on the Prize: Improved Biological Surface Registration via Forward Propagation

Many algorithms for surface registration risk producing significant errors if surfaces are significantly nonisometric. Manifold learning has been shown to be effective at improving registration quality, using information from an entire collection of surfaces to correct issues present in pairwise registrations. These methods, however, are not robust to changes in the collection of surfaces, or do not produce accurate registrations at a resolution high enough for subsequent downstream analysis. We propose a novel algorithm for efficiently registering such collections given initial correspondences with varying degrees of accuracy. By combining the initial information with recent developments in manifold learning, we employ a simple metric condition to construct a measure on the space of correspondences between any pair of shapes in our collection, which we then use to distill soft correspondences. We demonstrate that this measure can improve correspondence accuracy between feature points compared to currently employed, less robust methods on a diverse dataset of surfaces from evolutionary biology. We then show how our methods can be used, in combination with recent sampling and interpolation methods, to compute accurate and consistent homeomorphisms between surfaces.

preprint2020arXiv

STaR: Self-supervised Tracking and Reconstruction of Rigid Objects in Motion with Neural Rendering

We present STaR, a novel method that performs Self-supervised Tracking and Reconstruction of dynamic scenes with rigid motion from multi-view RGB videos without any manual annotation. Recent work has shown that neural networks are surprisingly effective at the task of compressing many views of a scene into a learned function which maps from a viewing ray to an observed radiance value via volume rendering. Unfortunately, these methods lose all their predictive power once any object in the scene has moved. In this work, we explicitly model rigid motion of objects in the context of neural representations of radiance fields. We show that without any additional human specified supervision, we can reconstruct a dynamic scene with a single rigid object in motion by simultaneously decomposing it into its two constituent parts and encoding each with its own neural representation. We achieve this by jointly optimizing the parameters of two neural radiance fields and a set of rigid poses which align the two fields at each frame. On both synthetic and real world datasets, we demonstrate that our method can render photorealistic novel views, where novelty is measured on both spatial and temporal axes. Our factored representation furthermore enables animation of unseen object motion.

preprint2020arXiv

SASSI -- Super-Pixelated Adaptive Spatio-Spectral Imaging

We introduce a novel video-rate hyperspectral imager with high spatial, and temporal resolutions. Our key hypothesis is that spectral profiles of pixels in a super-pixel of an oversegmented image tend to be very similar. Hence, a scene-adaptive spatial sampling of an hyperspectral scene, guided by its super-pixel segmented image, is capable of obtaining high-quality reconstructions. To achieve this, we acquire an RGB image of the scene, compute its super-pixels, from which we generate a spatial mask of locations where we measure high-resolution spectrum. The hyperspectral image is subsequently estimated by fusing the RGB image and the spectral measurements using a learnable guided filtering approach. Due to low computational complexity of the superpixel estimation step, our setup can capture hyperspectral images of the scenes with little overhead over traditional snapshot hyperspectral cameras, but with significantly higher spatial and spectral resolutions. We validate the proposed technique with extensive simulations as well as a lab prototype that measures hyperspectral video at a spatial resolution of $600 \times 900$ pixels, at a spectral resolution of 10 nm over visible wavebands, and achieving a frame rate at $18$fps.

preprint2020arXiv

Visualization of topology optimization designs with representative subset selection

An important new trend in additive manufacturing is the use of optimization to automatically design industrial objects, such as beams, rudders or wings. Topology optimization, as it is often called, computes the best configuration of material over a 3D space, typically represented as a grid, in order to satisfy or optimize physical parameters. Designers using these automated systems often seek to understand the interaction of physical constraints with the final design and its implications for other physical characteristics. Such understanding is challenging because the space of designs is large and small changes in parameters can result in radically different designs. We propose to address these challenges using a visualization approach for exploring the space of design solutions. The core of our novel approach is to summarize the space (ensemble of solutions) by automatically selecting a set of examples and to represent the complete set of solutions as combinations of these examples. The representative examples create a meaningful parameterization of the design space that can be explored using standard visualization techniques for high-dimensional spaces. We present evaluations of our subset selection technique and that the overall approach addresses the needs of expert designers.

preprint2020arXiv

Hybrid Function Representation for Heterogeneous Objects

Heterogeneous object modelling is an emerging area where geometric shapes are considered in concert with their internal physically-based attributes. This paper describes a novel theoretical and practical framework for modelling volumetric heterogeneous objects on the basis of a novel unifying functionally-based hybrid representation called HFRep. This new representation allows for obtaining a continuous smooth distance field in Euclidean space and preserves the advantages of the conventional representations based on scalar fields of different kinds without their drawbacks. We systematically describe the mathematical and algorithmic basics of HFRep. The steps of the basic algorithm are presented in detail for both geometry and attributes. To solve some problematic issues, we have suggested several practical solutions, including a new algorithm for solving the eikonal equation on hierarchical grids. Finally, we show the practicality of the approach by modelling several representative heterogeneous objects, including those of a time-variant nature.

preprint2021arXiv

SketchZooms: Deep multi-view descriptors for matching line drawings

Finding point-wise correspondences between images is a long-standing problem in image analysis. This becomes particularly challenging for sketch images, due to the varying nature of human drawing style, projection distortions and viewport changes. In this paper we present the first attempt to obtain a learned descriptor for dense registration in line drawings. Based on recent deep learning techniques for corresponding photographs, we designed descriptors to locally match image pairs where the object of interest belongs to the same semantic category, yet still differ drastically in shape, form, and projection angle. To this end, we have specifically crafted a data set of synthetic sketches using non-photorealistic rendering over a large collection of part-based registered 3D models. After training, a neural network generates descriptors for every pixel in an input image, which are shown to generalize correctly in unseen sketches hand-drawn by humans. We evaluate our method against a baseline of correspondences data collected from expert designers, in addition to comparisons with other descriptors that have been proven effective in sketches. Code, data and further resources will be publicly released by the time of publication.

preprint2021arXiv

Deep Preset: Blending and Retouching Photos with Color Style Transfer

End-users, without knowledge in photography, desire to beautify their photos to have a similar color style as a well-retouched reference. However, the definition of style in recent image style transfer works is inappropriate. They usually synthesize undesirable results due to transferring exact colors to the wrong destination. It becomes even worse in sensitive cases such as portraits. In this work, we concentrate on learning low-level image transformation, especially color-shifting methods, rather than mixing contextual features, then present a novel scheme to train color style transfer with ground-truth. Furthermore, we propose a color style transfer named Deep Preset. It is designed to 1) generalize the features representing the color transformation from content with natural colors to retouched reference, then blend it into the contextual features of content, 2) predict hyper-parameters (settings or preset) of the applied low-level color transformation methods, 3) stylize content to have a similar color style as reference. We script Lightroom, a powerful tool in editing photos, to generate 600,000 training samples using 1,200 images from the Flick2K dataset and 500 user-generated presets with 69 settings. Experimental results show that our Deep Preset outperforms the previous works in color style transfer quantitatively and qualitatively.

preprint2021arXiv

CARL: Controllable Agent with Reinforcement Learning for Quadruped Locomotion

Motion synthesis in a dynamic environment has been a long-standing problem for character animation. Methods using motion capture data tend to scale poorly in complex environments because of their larger capturing and labeling requirement. Physics-based controllers are effective in this regard, albeit less controllable. In this paper, we present CARL, a quadruped agent that can be controlled with high-level directives and react naturally to dynamic environments. Starting with an agent that can imitate individual animation clips, we use Generative Adversarial Networks to adapt high-level controls, such as speed and heading, to action distributions that correspond to the original animations. Further fine-tuning through the deep reinforcement learning enables the agent to recover from unseen external perturbations while producing smooth transitions. It then becomes straightforward to create autonomous agents in dynamic environments by adding navigation modules over the entire process. We evaluate our approach by measuring the agent's ability to follow user control and provide a visual analysis of the generated motion to show its effectiveness.

preprint2021arXiv

Neural Sparse Voxel Fields

Photo-realistic free-viewpoint rendering of real-world scenes using classical computer graphics techniques is challenging, because it requires the difficult step of capturing detailed appearance and geometry models. Recent studies have demonstrated promising results by learning scene representations that implicitly encode both geometry and appearance without 3D supervision. However, existing approaches in practice often show blurry renderings caused by the limited network capacity or the difficulty in finding accurate intersections of camera rays with the scene geometry. Synthesizing high-resolution imagery from these representations often requires time-consuming optical ray marching. In this work, we introduce Neural Sparse Voxel Fields (NSVF), a new neural scene representation for fast and high-quality free-viewpoint rendering. NSVF defines a set of voxel-bounded implicit fields organized in a sparse voxel octree to model local properties in each cell. We progressively learn the underlying voxel structures with a differentiable ray-marching operation from only a set of posed RGB images. With the sparse voxel octree structure, rendering novel views can be accelerated by skipping the voxels containing no relevant scene content. Our method is typically over 10 times faster than the state-of-the-art (namely, NeRF(Mildenhall et al., 2020)) at inference time while achieving higher quality results. Furthermore, by utilizing an explicit sparse voxel representation, our method can easily be applied to scene editing and scene composition. We also demonstrate several challenging tasks, including multi-scene learning, free-viewpoint rendering of a moving human, and large-scale scene rendering. Code and data are available at our website: https://github.com/facebookresearch/NSVF.

preprint2021arXiv

NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections

We present a learning-based method for synthesizing novel views of complex scenes using only unstructured collections of in-the-wild photographs. We build on Neural Radiance Fields (NeRF), which uses the weights of a multilayer perceptron to model the density and color of a scene as a function of 3D coordinates. While NeRF works well on images of static subjects captured under controlled settings, it is incapable of modeling many ubiquitous, real-world phenomena in uncontrolled images, such as variable illumination or transient occluders. We introduce a series of extensions to NeRF to address these issues, thereby enabling accurate reconstructions from unstructured image collections taken from the internet. We apply our system, dubbed NeRF-W, to internet photo collections of famous landmarks, and demonstrate temporally consistent novel view renderings that are significantly closer to photorealism than the prior state of the art.