EG2022

Permanent URI for this community

https://diglib.eg.org/handle/10.2312/2633163

Browse

Now showing 1 - 20 of 66

Neural Denoising for Spectral Monte Carlo Rendering
(The Eurographics Association, 2022) Rouphael, Robin; Noizet, Mathieu; Prévost, Stéphanie; Deleau, Hervé; Steffenel, Luiz-Angelo; Lucas, Laurent; Sauvage, Basile; Hasic-Telalovic, Jasminka
Spectral Monte Carlo (MC) rendering is still to be largely adopted partially due to the specific noise, called color noise, induced by wavelength-dependent phenomenons. Motivated by the recent advances in Monte Carlo noise reduction using Deep Learning, we propose to apply the same approach to color noise. Our implementation and training managed to reconstruct a noise-free output while conserving high-frequency details despite a loss of contrast. To address this issue, we designed a three-step pipeline using the contribution of a secondary denoiser to obtain high-quality results.
Interactive Facial Expression Editing with Non-linear Blendshape Interpolation
(The Eurographics Association, 2022) Roh, Ji Hyun; Kim, Seong Uk; Jang, Hanyoung; Seol, Yeongho; Kim, Jongmin; Pelechano, Nuria; Vanderhaeghe, David
The ability to manipulate facial animations interactively is vital for enhancing the productivity and quality of character animation. In this paper, we present a novel interactive facial animation editing system that can express the naturalness of non-linear facial movements in real-time. The proposed system is based on a fully automatic algorithm that maintains all positional constraints while deforming the facial mesh as realistic as possible. Our method is based on direct manipulation with non-linear blendshape interpolation. We formulate the facial animation editing as a two-step quadratic minimization and solve it efficiently. From our results, the proposed method produces the desired and realistic facial animation better compared to existing mesh deformation methods, which are mainly based on linear combination and optimization.
Advances in Neural Rendering
(The Eurographics Association and John Wiley & Sons Ltd., 2022) Tewari, Ayush; Thies, Justus; Mildenhall, Ben; Srinivasan, Pratul; Tretschk, Edith; Wang, Yifan; Lassner, Christoph; Sitzmann, Vincent; Martin-Brualla, Ricardo; Lombardi, Stephen; Simon, Tomas; Theobalt, Christian; Nießner, Matthias; Barron, Jon T.; Wetzstein, Gordon; Zollhöfer, Michael; Golyanik, Vladislav; Meneveaux, Daniel; Patanè, Giuseppe
Synthesizing photo-realistic images and videos is at the heart of computer graphics and has been the focus of decades of research. Traditionally, synthetic images of a scene are generated using rendering algorithms such as rasterization or ray tracing, which take specifically defined representations of geometry and material properties as input. Collectively, these inputs define the actual scene and what is rendered, and are referred to as the scene representation (where a scene consists of one or more objects). Example scene representations are triangle meshes with accompanied textures (e.g., created by an artist), point clouds (e.g., from a depth sensor), volumetric grids (e.g., from a CT scan), or implicit surface functions (e.g., truncated signed distance fields). The reconstruction of such a scene representation from observations using differentiable rendering losses is known as inverse graphics or inverse rendering. Neural rendering is closely related, and combines ideas from classical computer graphics and machine learning to create algorithms for synthesizing images from real-world observations. Neural rendering is a leap forward towards the goal of synthesizing photo-realistic image and video content. In recent years, we have seen immense progress in this field through hundreds of publications that show different ways to inject learnable components into the rendering pipeline. This state-of-the-art report on advances in neural rendering focuses on methods that combine classical rendering principles with learned 3D scene representations, often now referred to as neural scene representations. A key advantage of these methods is that they are 3D-consistent by design, enabling applications such as novel viewpoint synthesis of a captured scene. In addition to methods that handle static scenes, we cover neural scene representations for modeling nonrigidly deforming objects and scene editing and composition. While most of these approaches are scene-specific, we also discuss techniques that generalize across object classes and can be used for generative tasks. In addition to reviewing these state-ofthe- art methods, we provide an overview of fundamental concepts and definitions used in the current literature. We conclude with a discussion on open challenges and social implications.
Resolving Non-Manifoldness on Meshes from Dual Marching Cubes
(The Eurographics Association, 2022) Zint, Daniel; Grosso, Roberto; Gürtler, Philipp; Pelechano, Nuria; Vanderhaeghe, David
There are several methods that reconstruct surfaces from volume data by generating triangle or quad meshes on the dual of the uniform grid. Those methods often provide meshes with better quality than the famous marching cubes. However, they have a common issue: the meshes are not guaranteed to be manifold. We address this issue by presenting a post-processing routine that resolves all non-manifold edges with local refinement. New vertices are positioned on the trilinear interpolant. We verify our method on a wide range of data sets and show that we are capable of resolving all non-manifold issues.
Computational Assemblies: Analysis, Design, and Fabrication
(The Eurographics Association, 2022) Song, Peng; Wang, Ziqi; Livesu, Marco; Hahmann, Stefanie; Patow, Gustavo A.
Assemblies are ubiquitous in our daily life, such as toys, electronic devices, furniture, and architecture. They enable to build large and complex objects by composing small yet simpler parts, facilitating fabrication, storage, maintenance, and usage. However, designing assemblies is a highly non-trivial task because one needs to consider not only the properties of each individual components, but also of the whole assembly, such as aesthetics and stability. Motivated by recent advancements in digital fabrication, various computational techniques have been developed to analyze, design, and fabricate assemblies, aiming to enable general users to easily personalize them. This tutorial will give an introduction to these computational techniques, focusing on four fundamental aspects, i.e., parts fabricability, parts joining, assembly planning, and structural stability. In this tutorial, we will take a deep dive into computational methods to analyze these aspects for a given assembly as well as to design and fabricate assemblies that satisfy user-specified requirements in these aspects. This tutorial assumes knowledge of the fundamentals of computer graphics. Attendees should come away from this tutorial with a broad understanding of current work in computational assemblies, as well as familiarity with the necessary knowledge to start their own research in this area.
Inverse Computational Spectral Geometry
(The Eurographics Association, 2022) Rodolà, Emanuele; Cosmo, Luca; Ovsjanikov, Maks; Rampini, Arianna; Melzi, Simone; Bronstein, Michael; Marin, Riccardo; Hahmann, Stefanie; Patow, Gustavo A.
In the last decades, geometry processing has attracted a growing interest thanks to the wide availability of new devices and software that make 3D digital data available and manipulable to everyone. Typical issues faced by geometry processing algorithms include the variety of discrete representations for 3D data (point clouds, polygonal or tet-meshes and voxels), or the type of deformation this data may undergo. Powerful approaches to address these issues come from looking at the spectral decomposition of canonical differential operators, such as the Laplacian, which provides a rich, informative, robust, and invariant representation of the 3D objects. The focus of this tutorial is on computational spectral geometry. We will offer a different perspective on spectral geometric techniques, supported by recent successful methods in the graphics and 3D vision communities and older but notoriously overlooked results. We will discuss both the “forward” path typical of spectral geometry pipelines (e.g. computing Laplacian eigenvalues and eigenvectors of a given shape) with its widespread applicative relevance, and the inverse path (e.g. recovering a shape from given Laplacian eigenvalues, like in the classical “hearing the shape of the drum” problem) with its ill-posed nature and the benefits showcased on several challenging tasks in graphics and geometry processing.
EUROGRAPHICS 2022: Tutorials Frontmatter
(The Eurographics Association, 2022) Hahmann, Stefanie; Patow, Gustavo A.; Hahmann, Stefanie; Patow, Gustavo A.
Evaluating Bloom's Taxonomy-based Learning Modules for Parallel Coordinates Literacy
(The Eurographics Association, 2022) Peng, Ilena; Firat, Elif E.; Laramee, Robert S.; Joshi, Alark; Bourdin, Jean-Jacques; Paquette, Eric
In this paper, we present the results of an intervention designed to introduce parallel coordinates to students. The intervention contains six new modules inspired by Bloom's taxonomy that featured a combination of videos, tests, and tasks. We studied the impact of our modules with a corrective feedback mechanism inspired by Mastery Learning. Based on analyzing the data of our students, we found that students in the Corrective Immediate Feedback (CIF) group performed better on average on all the modules as compared to the students in the No Feedback (NF) group. In the tasks where students were required to construct parallel coordinates plots, students in the Corrective Immediate Feedback group produced plots with appropriate use of color, labels, legends, etc. Overall, students in both groups grew more confident in their ability to recognize parallel coordinates plots and expressed high confidence in their ability to interpret, create, and use parallel coordinates plots for data exploration and presentation in the future.
Multimodal Early Raw Data Fusion for Environment Sensing in Automotive Applications
(The Eurographics Association, 2022) Pederiva, Marcelo Eduardo; Martino, José Mario De; Zimmer, Alessandro; Sauvage, Basile; Hasic-Telalovic, Jasminka
Autonomous Vehicles became every day closer to becoming a reality in ground transportation. Computational advancement has enabled powerful methods to process large amounts of data required to drive on streets safely. The fusion of multiple sensors presented in the vehicle allows building accurate world models to improve autonomous vehicles' navigation. Among the current techniques, the fusion of LIDAR, RADAR, and Camera data by Neural Networks has shown significant improvement in object detection and geometry and dynamic behavior estimation. Main methods propose using parallel networks to fuse the sensors' measurement, increasing complexity and demand for computational resources. The fusion of the data using a single neural network is still an open question and the project's main focus. The aim is to develop a single neural network architecture to fuse the three types of sensors and evaluate and compare the resulting approach with multi-neural network proposals.
Graph Partitioning Algorithms for Rigid Body Simulations
(The Eurographics Association, 2022) Liu, Yinchu; Andrews, Sheldon; Pelechano, Nuria; Vanderhaeghe, David
We propose several graph partitioning algorithms for improving the performance of rigid body simulations. The algorithms operate on the graph formed by rigid bodies (nodes) and constraints (edges), producing non-overlapping and contiguous sub-systems that can be simulated in parallel by a domain decomposition technique. We demonstrate that certain partitioning algorithms reduce the computational time of the solver, and graph refinement techniques that reduce coupling between sub-systems, such as the Kernighan-Lin and Fiduccia-Mattheyses algorithms, give additional performance improvements.
Fast and Fine Disparity Reconstruction for Wide-baseline Camera Arrays with Deep Neural Networks
(The Eurographics Association, 2022) Barrios, Théo; Gerhards, Julien; Prévost, Stéphanie; Loscos, Celine; Sauvage, Basile; Hasic-Telalovic, Jasminka
Recently, disparity-based 3D reconstruction for stereo camera pairs and light field cameras have been greatly improved with the uprising of deep learning-based methods. However, only few of these approaches address wide-baseline camera arrays which require specific solutions. In this paper, we introduce a deep-learning based pipeline for multi-view disparity inference from images of a wide-baseline camera array. The network builds a low-resolution disparity map and retains the original resolution with an additional up scaling step. Our solution successfully answers to wide-baseline array configurations and infers disparity for full HD images at interactive times, while reducing quantification error compared to the state of the art.
Seamless Compressed Textures
(The Eurographics Association, 2022) Maggiordomo, Andrea; Tarini, Marco; Sauvage, Basile; Hasic-Telalovic, Jasminka
We present an algorithm to hide discontinuity artifacts at seams in GPU compressed textures. Texture mapping requires UV-maps, and UV-maps (in general) require texture seams; texture seams (in general) cause small visual artifacts in rendering; these can be prevented by careful, slight modifications a few texels around the seam. Unfortunately, GPU-based texture compression schemes are lossy and introduce their own slight modifications of texture values, nullifying that effort. The result is that texture compression may reintroduce the visual artefacts at seams. We modify a standard texture compression algorithm to make it aware of texture seams, resulting in compressed textures that still prevent the seam artefacts.
Consistent Multi- and Single-View HDR-Image Reconstruction from Single Exposures
(The Eurographics Association, 2022) Mohan, Aditya; Zhang, Jing; Cozot, Rémi; Loscos, Celine; Sauvage, Basile; Hasic-Telalovic, Jasminka
We propose a CNN-based approach for reconstructing HDR images from just a single exposure. It predicts the saturated areas of LDR images and then blends the linearized input with the predicted outputs. Two loss functions are used: the Mean Absolute Error and the Multi-Scale Structural Similarity Index. The choice of these loss functions allows us to outperform previous algorithms in the reconstructed dynamic range. Once the network trained, we input multi-view images to it to output multi-view coherent images.
View Dependent Decompression for Web-based Massive Triangle Meshes Visualisation
(The Eurographics Association, 2022) Cecchin, Alice; Du, Paul; Pastor, Mickaël; Agouzoul, Asma; Sauvage, Basile; Hasic-Telalovic, Jasminka
We introduce a framework extending an existing progressive compression-decompression algorithm for 3D triangular meshes. First, a mesh is partitioned. Each resulting part is compressed, then joined with one of its neighbours. These steps are repeated following a binary tree of operations, until a single compressed mesh remains. Decompressing the mesh involves progressively performing those steps in reverse, per node, and locally, by selecting the branch of the tree to explore. This method creates a compact and lossless representation of the model that allows its progressive and local reconstruction. Previously unprocessable meshes can be visualized on the web and mobile devices using this technique.
A Survey on Reinforcement Learning Methods in Character Animation
(The Eurographics Association and John Wiley & Sons Ltd., 2022) Kwiatkowski, Ariel; Alvarado, Eduardo; Kalogeiton, Vicky; Liu, C. Karen; Pettré, Julien; Panne, Michiel van de; Cani, Marie-Paule; Meneveaux, Daniel; Patanè, Giuseppe
Reinforcement Learning is an area of Machine Learning focused on how agents can be trained to make sequential decisions, and achieve a particular goal within an arbitrary environment. While learning, they repeatedly take actions based on their observation of the environment, and receive appropriate rewards which define the objective. This experience is then used to progressively improve the policy controlling the agent's behavior, typically represented by a neural network. This trained module can then be reused for similar problems, which makes this approach promising for the animation of autonomous, yet reactive characters in simulators, video games or virtual reality environments. This paper surveys the modern Deep Reinforcement Learning methods and discusses their possible applications in Character Animation, from skeletal control of a single, physically-based character to navigation controllers for individual agents and virtual crowds. It also describes the practical side of training DRL systems, comparing the different frameworks available to build such agents.
Learning Generic Local Shape Properties for Adaptive Super-Sampling
(The Eurographics Association, 2022) Reinbold, Christian; Westermann, Rüdiger; Pelechano, Nuria; Vanderhaeghe, David
We propose a novel encoder/decoder-based neural network architecture that learns view-dependent shape and appearance of geometry represented by voxel representations. Since the network is trained on local geometry patches, it generalizes to arbitrary models. A geometry model is first encoded into a sparse voxel octree of features learned by a network, and this model representation can then be decoded by another network in-turn for the intended task. We utilize the network for adaptive supersampling in ray-tracing, to predict super-sampling patterns when seeing coarse-scale geometry. We discuss and evaluate the proposed network design, and demonstrate that the decoder network is compact and can be integrated seamlessly into on-chip ray-tracing kernels. We compare the results to previous screen-space super-sampling strategies as well as non-network-based world-space approaches.
State-of-the-Art in the Architecture, Methods and Applications of StyleGAN
(The Eurographics Association and John Wiley & Sons Ltd., 2022) Bermano, Amit Haim; Gal, Rinon; Alaluf, Yuval; Mokady, Ron; Nitzan, Yotam; Tov, Omer; Patashnik, Or; Cohen-Or, Daniel; Meneveaux, Daniel; Patanè, Giuseppe
Generative Adversarial Networks (GANs) have established themselves as a prevalent approach to image synthesis. Of these, StyleGAN offers a fascinating case study, owing to its remarkable visual quality and an ability to support a large array of downstream tasks. This state-of-the-art report covers the StyleGAN architecture, and the ways it has been employed since its conception, while also analyzing its severe limitations. It aims to be of use for both newcomers, who wish to get a grasp of the field, and for more experienced readers that might benefit from seeing current research trends and existing tools laid out. Among StyleGAN's most interesting aspects is its learned latent space. Despite being learned with no supervision, it is surprisingly well-behaved and remarkably disentangled. Combined with StyleGAN's visual quality, these properties gave rise to unparalleled editing capabilities. However, the control offered by StyleGAN is inherently limited to the generator's learned distribution, and can only be applied to images generated by StyleGAN itself. Seeking to bring StyleGAN's latent control to real-world scenarios, the study of GAN inversion and latent space embedding has quickly gained in popularity. Meanwhile, this same study has helped shed light on the inner workings and limitations of StyleGAN. We map out StyleGAN's impressive story through these investigations, and discuss the details that have made StyleGAN the go-to generator. We further elaborate on the visual priors StyleGAN constructs, and discuss their use in downstream discriminative tasks. Looking forward, we point out StyleGAN's limitations and speculate on current trends and promising directions for future research, such as task and target specific fine-tuning.
Transparent Rendering and Slicing of Integral Surfaces Using Per-primitive Interval Arithmetic
(The Eurographics Association, 2022) Aydinlilar, Melike; Zanni, Cédric; Pelechano, Nuria; Vanderhaeghe, David
We present a method for efficient incorporation of integral surfaces within existing robust processing methods such as interval arithmetic and segment-tracing. We based our approach on high-level knowledge of the field function of the primitives. We show application to slicing and transparent rendering of integral surfaces based on interval arithmetic.
EUROGRAPHICS 2022: CGF 41-2 STARs Frontmatter
(The Eurographics Association and John Wiley & Sons Ltd., 2022) Meneveaux, Daniel; Patanè, Giuseppe; Meneveaux, Daniel; Patanè, Giuseppe
The Road to Vulkan: Teaching Modern Low-Level APIs in Introductory Graphics Courses
(The Eurographics Association, 2022) Unterguggenberger, Johannes; Kerbl, Bernhard; Wimmer, Michael; Bourdin, Jean-Jacques; Paquette, Eric
For over two decades, the OpenGL API provided users with the means for implementing versatile, feature-rich, and portable real-time graphics applications. Consequently, it has been widely adopted by practitioners and educators alike and is deeply ingrained in many curricula that teach real-time graphics for higher education. Over the years, the architecture of graphics processing units (GPUs) incrementally diverged from OpenGL's conceptual design. The more recently introduced Vulkan API provides a more modern, fine-grained approach for interfacing with the GPU. Various properties of this API and overall trends suggest that Vulkan could soon replace OpenGL in many areas. Hence, it stands to reason that educators who have their students' best interests at heart should provide them with corresponding lecture material. However, Vulkan is notoriously verbose and rather challenging for first-time users, thus transitioning to this new API bears a considerable risk of failing to achieve expected teaching goals. In this paper, we document our experiences after teaching Vulkan in an introductory graphics course side-by-side with conventional OpenGL. A final survey enables us to draw conclusions about perceived workload, difficulty, and students' acceptance of either approach and identify suitable conditions and recommendations for teaching Vulkan to undergraduate students.

Browse

Browsing EG2022 by Issue Date

Results Per Page

Sort Options