Search

3D meshes are commonly utilized and may be considered to be the most popular surface representation in computer graphics due to their simplicity, efficiency and flexibility. However, the explicit storage of mesh vertices and connectivity, as in widely-used PLY and OBJ file formats, leads to substantial memory consumption. This, in turn, directly affects the processing and transmission in downstream applications. Though mesh simplification and mesh compression are common strategies to lessen memory consumption, they exhibit inherent limitations either in maintaining a balance between accuracy, efficiency, memory usage and mesh quality, or breaking the simplicity of explicit storage and struggling with optimizing the trade-off between compression performance and computational resource consumption. To overcome these limitations, inspired by the Gaussian Mixture Model (GMM), this paper proposes a GMM vertex-optimization-based implicitly-connected quadrilateral format for 3D mesh storage, named GMM-ICQ. Extensive qualitative and quantitative evaluations demonstrate that the GMM-ICQ format achieves efficient compression by retaining only a small amount of vertex information, while preserving sharp features and maintaining relatively high mesh quality. It also exhibits a certain degree of robustness in the presence of noise interference. Furthermore, benefiting from the inherent grid-based connectivity, the GMM-ICQ format maintains the simplicity of explicit storage and can be implemented as a progressive variant without incurring additional computational overhead. [Display omitted] • We present a GMM vertex-optimization-based implicitly-connected quadrilateral format for 3D mesh storage. • Simultaneously balances accuracy, efficiency, memory usage, and mesh quality. • Preserves the simplicity of explicit storage (such as PLY and OBJ). • No additional computational overhead needed for progressive variant implementation. [ABSTRACT FROM AUTHOR]

Immersive Virtual Reality (IVR) provides a platform where the real world can be replicated to a point where users can act and react in the virtual world as they would in reality. However, rendering visual stimuli is computationally heavy. Thus, optimizations must be done to take advantage of computational systems by studying our perception of reality. This study investigated parameters related to light rendering (Global Illumination, Ambient Occlusion, Screen Space Reflections (SSR) and Direct Shadows) in real-time in a virtual replica of a real place using IVR. Participants experienced both virtual and real rooms with only one flashlight and changed the quality settings of the considered parameters so that their sense of reality would be the closest to the one they felt when they experienced the real room. Participants were given a budget to drive them to prioritize what parameters, and their level of quality, are the most important for their sense of reality. Results indicated that participants considered Global Illumination the most important factor, closely followed by Direct Shadows. Ambient Occlusion and Reflections (Screen Space Reflections) were the less prioritized parameters. We conclude that in a lighting setting where only dynamic lights are used, Global Illumination and Direct Shadows should be prioritized over SSR Reflections and Ambient Occlusion when computational power is limited. [Display omitted] • GI & Direct Shadows crucial for realism, while AO & Reflections less significant. • GI impact: none to low-quality > low- to high-quality. • Users insisted in having Direct Shadows at highest quality. • Perception of reality varies widely even when comparing directly with a real scene. [ABSTRACT FROM AUTHOR]

The curve skeleton is an important shape descriptor which has been utilized in various applications in computer graphics, machine vision, and artificial intelligence. In this study, the endpoint-based part-aware curve skeleton (EPCS) extraction method for low-quality point clouds is proposed. The novel random center shift (RCS) method is first proposed for detecting the endpoints on point clouds. The endpoints are used as the initial seed points for dividing each part into layers, and then the skeletal points are obtained by computing the center points of the oriented bounding box (OBB) of the layers. Subsequently, the skeletal points are connected, thus forming the branches. Furthermore, the multi-vector momentum-driven (MVMD) method is also proposed for locating the junction points which connect the branches. Due to the shape differences between different parts on point clouds, the global topology of the skeleton is finally optimized by removing the redundant junction points, re-connecting some branches using the proposed MVMD method, and applying an interpolation method based on the splitting operator. Consequently, a complete and smooth curve skeleton is achieved. The proposed EPCS method is compared with several state-of-the-art methods, and the experimental results verify its robustness and effectiveness. Furthermore, the skeleton extraction and model segmentation results on challenging point clouds of broken Terracotta also highlight the utility of the proposed method. [Display omitted] • A novel EPCS method is proposed for curve skeleton extraction, which is robust to noises, outliers, and missing data, and it can be applied to low-quality point clouds. The proposed EPCS requires no preprocessing, such as mesh reconstruction, Voronoi graph generation, or the computation of geometric invariants. • A novel RCS method is proposed for the detection of the endpoints on point clouds, which can prevent missing branches during skeleton extraction. • The MVMD method is proposed for locating the junction points that connect the branches, significantly improving the completeness and smoothness of the skeletal curves in the connected regions. • The proposed EPCS outperforms current state-of-the-art methods with respect to effectiveness, robustness, and efficiency. [ABSTRACT FROM AUTHOR]

Diffusion magnetic resonance imaging quantifies the diffusion of water molecules. It is unique in showing white matter fiber structures and connectivity in-vivo, relying on the fact that the water molecules' displacements are more significant along with the fiber orientation than in other directions due to biological barriers. High angular resolution diffusion imaging (HARDI) acquisition enables the differentiation of multiple fiber signals passing through a voxel. Advanced methods for processing diffusion data were developed to synthesize the HARDI data of each voxel into an orientation distribution function (ODF) that relates each direction sample to its probability of water diffusion. Visualizing ODFs through glyphs helps gain insight into the local structure of the brain white matter. Because of the high dimensionality of the data, interactive glyph exploration remains a challenge for a volume containing millions of voxels. This work presents a 3D triangle-based multi-resolution rendering scheme for the ODF glyph. We explored data properties and GPU features to make the triangle-based rendering interactive. Linear performance relative to some fixed-resolution rendered glyphs was assessed using NVIDIA® Nsight Visual Studio T M. A control volume, available in the Retest dataset of the WU-Minn Human Connectome Project, was used to assess the effectiveness of interactive 3D visualization over a multiplanar and curvilinear reformatted volume of ODF glyphs in voxels with multiple fibers, namely the triple crossing at the centrum semiovale, the fanning at the corona radiata, and the crossing of the cingulum and corpus callosum fiber bundles. [Display omitted] [ABSTRACT FROM AUTHOR]

Raw point clouds captured by sensing devices are often contaminated with noise, which perturbs the fidelity of the original geometric information. Point cloud denoising is therefore an inseparable post-processing step, aiming to remove the noise in the point clouds. Existing point cloud denoising approaches are typically trained on datasets that have uniform point distributions and densities, making them unsuitable for effectively denoising point clouds with severe noise or irregular point distributions. In this paper, we introduce a novel random screening-based feature aggregation method for point cloud denoising. Our key insight is that merging features of dense and sparse points assists with enhancing the quality of point cloud denoising results. In specific, our approach involves randomly screening the features of local point patches and fusing richer geometric information of denser points into sparser point representations. Comprehensive experiments demonstrate that our method achieves state-of-the-art performance in the point cloud denoising task on both synthetic and real-world datasets. [Display omitted] • Random screening-based point feature aggregation is useful for point cloud denoising. • Merging features of dense points into sparser points boosts point cloud denoising. • Our pipeline demonstrates robustness to noise and saves processing time. • Our technique achieves impressive results on severe noise and irregular points. [ABSTRACT FROM AUTHOR]

The retrieval of 3D objects has gained significant importance in recent years due to its broad range of applications in computer vision, computer graphics, virtual reality, and augmented reality. However, the retrieval of 3D objects presents significant challenges due to the intricate nature of 3D models, which can vary in shape, size, and texture, and have numerous polygons and vertices. To this end, we introduce a novel SHREC challenge track that focuses on retrieving relevant 3D animal models from a dataset using sketch queries and expedites accessing 3D models through available sketches. Furthermore, a new dataset named ANIMAR was constructed in this study, comprising a collection of 711 unique 3D animal models and 140 corresponding sketch queries. Our contest requires participants to retrieve 3D models based on complex and detailed sketches. We receive satisfactory results from eight teams and 204 runs. Although further improvement is necessary, the proposed task has the potential to incentivize additional research in the domain of 3D object retrieval, potentially yielding benefits for a wide range of applications. We also provide insights into potential areas of future research, such as improving techniques for feature extraction and matching and creating more diverse datasets to evaluate retrieval performance. • ANIMAR dataset was constructed with 711 3D animal models and 140 sketch queries. • SHREC challenge track focusing on Text-based 3D ANIMAl model fine-grained Retrieval. • We receive satisfactory results from eight teams and 204 runs. [ABSTRACT FROM AUTHOR]

The realm of 3D computer vision and graphics has experienced exponential growth recently, enabling the creation of realistic virtual environments and digital representations of real-world objects. Central to this progression are 3D reconstruction methods that facilitate the virtualization of shape, color, and surface details of real objects. Current methods predominantly employ neural scene representations, which despite their efficacy, grapple with limitations such as necessitating a high number of captured images and the complexity of transforming these representations into explicit geometric forms. An alternative strategy that has gained traction is the deployment of methods such as physically-based differentiable rendering (PBDR) and inverse rendering. These approaches require fewer viewpoints, yield explicit format results, and ensure a smoother transition to other representation methods. To meaningfully assess the performance of different 3D reconstruction methods, it is imperative to utilize benchmark scenes for comparison. Despite the existence of standard objects and scenes within the literature, there is a noticeable deficiency in real-world benchmark data that concurrently captures camera, illumination, and scene parameters — all critical to high-fidelity 3D reconstructions using PBDR and inverse rendering-based methods. In this research, we introduce a methodology for capturing real-world scenes as virtual scenes, integrating illumination parameters alongside camera and scene parameters to enhance the veracity of virtual representations. In addition, we introduce a set of ten real-world scenes, along with their virtual counterparts, designed as benchmarks. These benchmarks encompass a fundamental variety of geometric constructs, including convex, concave, plain, and mixed surfaces. Additionally, we demonstrate the 3D reconstruction results of state-of-the-art 3D reconstruction methods employing PBDR in real-world scenes, using both established methodologies and our proposed one. • Recent advancements in computer vision and graphics enabled realistic virtual scenes. • Cutting-edge computer graphics techniques can model a given virtual environment. • The latest virtual scene optimization methods can be applied to real-world scenes. • Combining multiscale L1 loss with structural losses improves reconstruction accuracy. [Display omitted] [ABSTRACT FROM AUTHOR]

We present SpecTrHuMS, a Spectral Transformer for 3D triangular Human Mesh Sequence learning which combines known deep learning models with spectral mesh processing to capture characteristics of 3D shapes as well as temporal dependencies between the frames. Unlike previous works in this field, our approach is able to work directly with a compressed representation of the geometry, the spectral coefficients, rather than relying solely on skeleton joints that does not contain surface information. The vertices of each mesh of a sequence are first projected on the eigenvectors of the Graph Laplacian computed from the common triangulation. A convolutional encoder then encodes each frame into lower dimensional latent variables that preserve as much as possible the spectral information. These latent variables are next passed through a transformer architecture so that the model understands the context of the sequence and learns temporal dependencies between the frames. Each frame of the transformer's output is then decoded by a convolutional decoder which aims to reconstruct the input spectral coefficients. Finally, all frames are transformed back into the spatial domain, resulting in a general process able to treat 4D surfaces with a constant connectivity. Our method is evaluated on a prediction task on AMASS, a dataset of human surface sequences, showing the ability of our model to produce realistic movements while preserving the identity of a subject, and showing that this work is a significant step towards efficient and high-quality representation of triangular mesh sequences with constant connectivity. Additional experiments show that our model can be easily extended to other tasks such as long term prediction, completion and that it is generalizable to other datasets with constant connectivity. This work opens up new possibilities for applications in the fields of animation, virtual reality, and computer graphics. Pretrained models, the code to train them and the code to create datasets will be made publicly available. [Display omitted] • This work focuses on generating human triangular mesh sequences. • It consists in the association of a convolutional autoencoder and a transformer. • Spectral coefficients computed from the Graph Laplacian are given as input. • The context of the sequence is understood using a transformer. • The trained model is able to predict movements and preserve the identity of subjects. [ABSTRACT FROM AUTHOR]

Digital color restitution aims to digitally restore the original colors of a painting. Existing image editing applications can be used for this purpose, but they require a select-and-edit workflow and thus they do not scale well to large collections of paintings or different regions of the same painting. To address this issue, we propose an automated workflow that requires only a few representative source colors and associated target colors as input from art historians. The system then creates a control grid to model a deformation of the CIELAB color space. Such deformation can be applied to arbitrary images of the same painting. The proposed approach is suitable for restituting the color of images from a large photographic campaign, as well as for the textures of 3D reconstructions of a monument. We demonstrate the benefits of our method on a collection of mural paintings from a medieval monument. • Automated workflow for digital color restitution. • Deformation of CIELAB color space to test art historian hypotheses. • Applicable in photographic campaigns and 3D reconstruction. [ABSTRACT FROM AUTHOR]

The computation of power diagrams (or weighted Voronoi diagrams) is a fundamental task in computational geometry and computer graphics. To accomplish the computation, we provide a different way from the existing ones for lifting the weighted seeds to a set of points in the space of one dimension higher, then the power cells can be directly obtained by computing the intersections of the Voronoi cells of these lifting points and the original space. This property enables us to apply the method based on the k -nearest neighbors query to the generation of power diagrams. Each power cell is obtained by sequentially clipping the input domain using the bisectors between its seed and the k -nearest neighbors. Experimental results demonstrate that our method outperforms the state-of-the-art one based on regular triangulation in terms of efficiency for general cases in the 2D and 3D spaces. [Display omitted] • We provide a variant for lifting the weighted seeds to higher dimensional points. • We propose an efficient method for power diagram computation based on the kNN query. [ABSTRACT FROM AUTHOR]

In recent years, the computer vision and computer graphics communities have seen the emergence of new classes of models for representing objects based on Neural Radiance Fields (NeRFs). These techniques use ideas based on traditional methods from computer graphics, for example, radiance fields, and combine them with implicit neural representations and neural image-based rendering (IBR). However, one significant drawback of this class of techniques was that they were too slow, taking in the range of hours in high-end GPUs. Due to these limitations, new techniques have been created for the fast reconstruction of scenes, such as Direct Voxel Grid Optimization (DirectVoxGO). Alongside this limitation, initially, NeRFs had limited compositing and modeling capabilities. For example, a simple task such as separating the background from the foreground could not be modeled with NeRFs until the emergence of new techniques such as NeRF++. Our method extends DirectVoxGO to allow the handling of unbounded scenes inspired by some ideas from NeRF++ and also adapts it to incorporate elements from a neural hashing approach employed by other works. Our technique improved photorealism compared with DirectVoxGO and Plenoxels on a subset of the Light Field Dataset on average in at least 2%, 8%, and 8% for Peak Signal-to-Noise Ratio (PSNR), Structural Similarity Index Measure (SSIM), and Learned Perceptual Image Patch Similarity (LPIPS) metrics, respectively, while also being an order of magnitude faster than NeRF++. Also, we demonstrate that, for the evaluated scenes, our technique has comparable training time and memory consumption than previous works. Code is available in https://github.com/danperazzo/dvgoplusplus. [Display omitted] • Define and develop an improvement of DirectVoxGO for scenes and extraction of a 3D model of an object of interest. • Perform qualitative evaluations of our method on widely-used datasets. • Perform quantitative evaluations of our method on these widely-used datasets. [ABSTRACT FROM AUTHOR]

Reconstructing the shape of hand-held objects from single-view color images is a long-standing problem in computer vision and computer graphics. The task is complicated by the ill-posed nature of single-view reconstruction, as well as potential occlusions due to both the hand and the object. Previous works mostly handled the problem by utilizing known object templates as priors to reduce the complexity. In contrast, our paper proposes a novel approach without knowing the object templates beforehand but by exploiting prior knowledge of contacts in hand-object interactions to train an attention-based network that can perform precise hand-held object reconstructions with only a single forward pass in inference. The network we propose encodes visual features together with contact features using a multi-head attention module as a way to condition the training of a neural field representation. This neural field representation outputs a Signed Distance Field representing the reconstructed object and extensive experiments on three well-known datasets demonstrate that our method achieves superior reconstruction results even under severe occlusion compared to the state-of-the-art techniques. • Conditioning on contact, we reconstruct the hand-held object from single-view images. • We use an end-to-end attention-based network to better encode contact priors. • We achieve state-of-the-art results on Obman, HO3D, and MOW datasets. [ABSTRACT FROM AUTHOR]

Shape matching is a fundamental problem in computer graphics with many applications. Functional maps translate the point-wise shape-matching problem into its functional counterpart and have inspired numerous solutions over the last decade. Nearly all the solutions based on functional maps rely on the eigenfunctions of the Laplace–Beltrami Operator (LB) to describe the functional spaces defined on the surfaces and then convert the functional correspondences into point-wise correspondences. However, this final step is often error-prone and inaccurate in tiny regions and protrusions, where the energy of LB does not uniformly cover the surface. We propose a new functional basis Principal Components of a Dictionary (PC D) to address such intrinsic limitation. PC D constructs an orthonormal basis from the Principal Component Analysis (PCA) of a dictionary of functions defined over the shape. These dictionaries can target specific properties of the final basis, such as achieving an even spreading of energy. Our experimental evaluation compares seven different dictionaries on established benchmarks, showing that PC D is suited to target different shape-matching scenarios, resulting in more accurate point-wise maps than the LB basis when used in the same pipeline. This evidence provides a promising alternative for improving correspondence estimation, confirming the power and flexibility of functional maps. [Display omitted] • A procedure to adaptively define basis for functional spaces over discrete 3D meshes • A general method that applies to different redundant dictionaries of functions • Achieve state-of-the-art shape matching accuracy in a functional map framework • Disclose the impact of more even distribution of the energy of a basis in matching • Quantitative assessment of seven bases in comparison to existing alternatives [ABSTRACT FROM AUTHOR]

New technology such as virtual reality can help computer graphics education, for example, by providing the opportunity to illustrate challenging 3D procedures. RePiX VR is a virtual reality tool for computer graphics education that focuses on teaching the core ideas of the rendering pipeline. This paper describes the development and two initial evaluations, which aimed to strengthen the usability, review requirements for different stakeholders, and build infrastructure for learning analytics and research. The integration of learning analytics raises the question of appropriate indicators to be approached through exploratory data analysis. In addition to learning analytics, the evaluation includes quantitative techniques to get insights about usability, and didactical feedback. This paper discusses advanced aspects of learning in VR and looks specifically at movement behavior. According to the evaluations, even learners without prior experience can utilize the VR tool to pick up the fundamentals of computer graphics. [Display omitted] • Evaluated educational VR environment for teaching Computer Graphics. • Teaching Computer Graphics in Virtual Reality is promising. • Learners have various movement and teleportation pattern and different interaction behavior. • A comparison of Desktop and VR users shows differences between groups, as well as a comparison of novices and experts. • The evaluation contains Multimodal Learning Analytics, Quantitative Feedback, and Usability aspects. [ABSTRACT FROM AUTHOR]

Weathering phenomena are ubiquitous in urban environments, where it is easy to observe severely degraded old buildings as a result of water penetration. Despite being an important part of any realistic city, this kind of phenomenon has received little attention from the Computer Graphics community compared to stains resulting from biological or flow effects on the building exteriors. In this paper, we present physically-inspired deep weathering effects, where the penetration of humidity (i.e., water particles) and its interaction with a building's internal structural elements result in large, visible degradation effects. Our implementation is based on a particle-based propagation model for humidity propagation, coupled with a spring-based interaction simulation that allows chemical interactions, like the formation of rust, to deform and destroy a building's inner structure. To illustrate our methodology, we show a collection of deep degradation effects applied to urban models involving the creation of rust or of ice within walls. [Display omitted] • Physically-inspired simulation of the penetration of water particles in walls, dealing with multiple types of materials. • Approximate simulation of the interactions of water with the inner structures of the wall. • Expansion and deformation, such as water into ice, rusting, or plaster deformation. • Simulation of the formation of cracks inside the wall, leading to broken concrete, loose plaster, etc. • Cracks can accelerate previous steps if increased exposure (e.g., rain). [ABSTRACT FROM AUTHOR]

Building on our original Virtual Ray Tracer tool, we present Virtual Ray Tracer 2.0 , an interactive and gamified application that allows students/users to view and explore the ray tracing process in real-time. The application shows a scene containing a camera casting rays which interact with objects in the scene. Users are able to modify and explore ray properties such as their animation speed, the number of rays and their visual style, as well as the material properties of the objects in the scene. The goal of the application is to help the users – students of Computer Graphics and the general public – to better understand the ray tracing process and its characteristics. This includes not only the basics of ray tracing, but also more advanced concepts such as soft shadows. To invite users to learn and explore, various explanations and scenes are provided by the application at different levels of complexity, each with a step-by-step tutorial. Several user studies showed the effectiveness of the tool in supporting the understanding and teaching of ray tracing. The educational tool is built with the cross-platform engine Unity, and we make it fully available to be extended and/or adjusted to fit the requirements of courses at other institutions, educational tutorials, or of enthusiasts from the general public. [Display omitted] • Area lights and soft shadows (via super/area sampling). • Novel ray visualization styles (such as according to importance and/or color contribution). • Bounding volumes for acceleration (dynamic axis-aligned bounding boxes and octrees). • Gamification (with tailored tutorials per level/scene, reward badges, and more). • A mobile and web version of the tool. [ABSTRACT FROM AUTHOR]

Simulating trajectories of virtual crowds is a commonly encountered task in Computer Graphics. Several recent works have applied Reinforcement Learning methods to animate virtual agents, however they often make different design choices when it comes to the fundamental simulation setup. Each of these choices comes with a reasonable justification for its use, so it is not obvious what is their real impact, and how they affect the results. In this work, we analyze some of these arbitrary choices in terms of their impact on the learning performance, as well as the quality of the resulting simulation measured in terms of the energy efficiency. We perform a theoretical analysis of the properties of the reward function design, and empirically evaluate the impact of using certain observation and action spaces on a variety of scenarios, with the reward function and energy usage as metrics. We show that directly using the neighboring agents' information as observation generally outperforms the more widely used raycasting. Similarly, using nonholonomic controls with egocentric observations tends to produce more efficient behaviors than holonomic controls with absolute observations. Each of these choices has a significant, and potentially nontrivial impact on the results, and so researchers should be mindful about choosing and reporting them in their work. [Display omitted] • Crowd simulation details affect the results when trained with reinforcement learning. • It is typically beneficial for agents to directly observe others' positions. • Egocentric observations and actions are better than global ones. • Reward design is highly impactful and nontrivial. • Simple reward functions do not produce human-like behavior. [ABSTRACT FROM AUTHOR]

Generating 3D models from 2D images or sketches is a widely studied important problem in computer graphics. We describe the first method to generate a 3D human model from a single sketched stick figure. In contrast to the existing human modeling techniques, our method does not require a statistical body shape model. We exploit Variational Autoencoders to develop a novel framework capable of transitioning from a simple 2D stick figure sketch, to a corresponding 3D human model. Our network learns the mapping between the input sketch and the output 3D model. Furthermore, our model learns the embedding space around these models. We demonstrate that our network can generate not only 3D models, but also 3D animations through interpolation and extrapolation in the learned embedding space. In addition to 3D human models, we produce 3D horse models in order to show the generalization ability of our framework. Extensive experiments show that our model learns to generate compatible 3D models and animations with 2D sketches. [Display omitted] • Our method generates 3D models and animations from given 2D stick figure sketches. • We exploit Variational Autoencoders in our framework for 2D sketch-3D model mapping. • Our method does not require any other input such as a rigged template model. [ABSTRACT FROM AUTHOR]

Developable surfaces are widely used in Computer-Aided Design (CAD), computer graphics, architecture geometry and manufacturing. In this paper, we present an approach to segmentation and approximation of an input triangular mesh with developable patches. First, exact developable regions are detected and extracted from the input mesh. Then the non-developable region is further segmented and approximated also by developable patches. The parametric equations for all obtained developable patches are simultaneously computed, whose iso-parameter curves can serve directly as the milling paths in CNC machining. Examples and comparisons with existing works are provided. [Display omitted] • A novel method for segmenting and approximating an input mesh with developable patches is proposed. • The exact parametric equations for all segment patches are computed. • A new linearity and smoothness measures to detect curve-like features on Gauss Images is provided. [ABSTRACT FROM AUTHOR]

The Shape Modeling International awards (SMI awards) were introduced to commemorate the passing of SMI founder, Professor Kunii. Since 2021, the SMI awards recognize exceptional contributors to Shape Modeling. Currently, there are three awards: the Tosiyasu Kunii Distinguished Researcher, the Young Investigator, and the Alexander Pasko Service Award. The 2024 Distinguished Researcher awardees are Gershon Elber and Stefanie Hahmann. The 2024 Young Investigators are Gianmarco Cherchi and Amal Dev Parakkat. The 2024 Service Awardee is Ergun Akleman. This article provides interviews with the five SMI 2024 award winners. [Display omitted] • We present interviews with Shape Modeling International (SMI) 2024 award winners. • We provided an overview of history of SMI conference. • For each award winner, we provided a bio and interview. • We also cited their papers published in proceedings of Shape Modeling International and Computers & Graphics to demonstrate their contributions to SMI community. • We provided also rare pictures from SMI conferences. [ABSTRACT FROM AUTHOR]

Search and retrieval remains a major research topic in several domains, including computer graphics, computer vision, engineering design, etc. A search engine requires primarily an input search query and a database of items to search from. In engineering, which is the primary context of this paper, the database consists of 3D CAD models, such as washers, pistons, connecting rods, etc. A query from a user is typically in the form of a sketch, which attempts to capture the details of a 3D model. However, sketches have certain typical defects such as gaps, over-drawn portions (multi-strokes), etc. Since the retrieved results are only as good as the input query, sketches need cleaning-up and enhancement for better retrieval results. In this paper, a deep learning approach is proposed to improve or clean the query sketches. Initially, sketches from various categories are analysed in order to understand the many possible defects that may occur. A dataset of cleaned-up or enhanced query sketches is then created based on an understanding of these defects. Consequently, an end-to-end training of a deep neural network is carried out in order to provide a mapping between the defective and the clean sketches. This network takes the defective query sketch as the input and generates a clean or an enhanced query sketch. Qualitative and quantitative comparisons of the proposed approach with other state-of-the-art techniques show that the proposed approach is effective. The results of the search engine are reported using both the defective and enhanced query sketches, and it is shown that using the enhanced query sketches from the developed approach yields improved search results. [Display omitted] • The first learning-based strategy to clean rough query sketches of 3D CAD models • Introduces SketchCleanNet — an end-to-end image translation scheme • SketchCleanNet aims to understand the mapping between rough sketches and clean query images • A novel scheme to calculate the loss is introduced • Dataset Contribution: The resulting enhanced query sketch dataset is made available publicly. • This paper will significantly contribute to the research community and give researchers opportunities to develop new algorithms for search and retrieval of 3D mechanical components. [ABSTRACT FROM AUTHOR]

This paper tackles the volumetric representation of geophysical and geotechnical data, gathered during exploration surveys of the subsoil; in particular, we focus on the modeling and analysis of underwater deposits. The creation of a 3D model as support to geological interpretation has to take into account the heterogeneity of the input data, coming from offshore acquisition campaigns. Some data are massive, but cover the domain unevenly, e.g., along dense differently spaced lines, while others are very sparse, e.g., borehole locations with soil sampling and CPTU (Piezocone Penetration Test) locations. A automatic process is presented to generate the subsurfaces and volume defining a sub-seabed deposit, starting from the identification of relevant morphological features in seismic data. In particular, simplification and refinement based on geostatistics have been applied to generate regular 2D meshes from strongly anisotropic data, in order to improve the quality of the final 3D tetrahedral mesh. Furthermore, we also use geostatistics to predict geotechnical parameters from local surveys and estimate their distribution on the whole domain: in this way the 3D model will include relevant geological features of the deposit and allow extrapolating different geotechnical information with associated uncertainty. The volume characterization and its 3D inspection will support geological analysis and planning of future engineering activities. The developed methodology has been tested on two real case studies. • We implemented a complete pipeline for the 3D modeling of stratified deposits • Our research action follows the surface-based approach • We combine computer graphics methodologies with geophysics and geostatistics [ABSTRACT FROM AUTHOR]

The realistic reconstruction of real-world trees is a challenging task in the community of computer graphics because natural trees have complex structures of branches and leaves. Existing terrestrial laser scanning (TLS) system is able to capture dense and precise tree point clouds, yet the TLS system is expensive and not easy to carry around. An alternative low-cost and portable way is the reconstruction of tree point cloud from multiple view images. However, it is usually difficult to reconstruct a complete tree point cloud because of the texture similarity of branches and leaves as well as the lack of a sufficient number of images. Thus, we propose a new approach for reconstructing tree point clouds and geometries from sparse images. We first infer the camera parameters of each image, and then calculate the bounding volume of a tree from the camera parameters. Next, we set the mask of each image and the resolution of voxel, and then project each voxel in 3D space to all the mask images to determine the validity of the voxel. To alleviate the miss deletion of valid voxel, we utilize a boundary threshold and adjust mask resolution for robust point cloud reconstruction. Finally, an efficient tree reconstruction method is proposed to generate plausible tree geometries. We tested 6 different tree species that contain deciduous and evergreen trees, and the results showed that our approach is able to generate complete tree point cloud and realistic tree models even from a few number of images. [Display omitted] • A low-cost volumetric reconstruction approach for generating dense tree point cloud from sparse images. • An efficient tree reconstruction method to generate realistic tree model from the tree point cloud. • Polygonal tree models can be exported for rendering or content creation. [ABSTRACT FROM AUTHOR]

Time-varying connectivity of triangle mesh sequences leads to substantial difficulties in their processing. Unlike editing sequences with constant connectivity, editing sequences with varying connectivity requires addressing the problem of temporal correspondence between the frames of the sequence. We present a method for time-consistent editing of triangle mesh sequences with varying connectivity using sparse temporal correspondence, which can be obtained using existing methods. Our method includes a deformation model based on the usage of the sparse temporal correspondence, which is suitable for the temporal propagation of user-specified deformations of the edited surface with respect to the shape and true topology of the surface while preserving the individual connectivity of each frame. Since there is no other method capable of comparable types of editing on time-varying meshes, we compare our method and the proposed deformation model with a baseline approach and demonstrate the benefits of our framework. [Display omitted] • This paper presents a method for editing mesh sequences with varying connectivity. • The proposed method exploits sparse temporal correspondences. • Dual quaternions are used to represent the transformations. [ABSTRACT FROM AUTHOR]

Point set registration has been actively studied in computer vision and graphics. Optimization algorithms are at the core of solving registration problems. Traditional optimization approaches are mainly based on the gradient of objective functions. The derivation of objective functions makes it challenging to find optimal solutions for complex optimization models, especially for those applications where accuracy is critical. Learning-based optimization is a novel approach to address this problem, which learns the gradient direction from datasets. However, many learning-based optimization algorithms learn gradient directions via a single feature extracted from the dataset, which will cause the updating direction to be vulnerable to perturbations around the data, thus falling into a bad stationary point. This paper proposes the General Discriminative Optimization (GDO) method that updates a gradient path automatically through the trade-off among contributions of different features on updating gradients. We illustrate the benefits of GDO with tasks of 3D point set registrations and show that GDO outperforms the state-of-the-art registration methods in terms of accuracy and robustness to perturbations. [Display omitted] • Cast the point sets registration as a learning-based optimization problem. • Utilize features of point sets to update gradients without the Hessian matrix. • Collaborate the different features to reduce the influence of perturbations. • Outperform the other advanced registration methods on accuracy and robustness. [ABSTRACT FROM AUTHOR]

We present a new approach for path finding in weighted graphs using pre-computed minimal distance fields. By selecting the most promising minimal distance field at any given node and switching between them, our algorithm aims to find the shortest path possible. As we show, this approach scales excellently for various topologies, graph sizes and hardware specifications while maintaining a mean length error below 1% and reasonable memory consumption. By utilizing a simplified structure and keeping backtracking to a minimum, we are able to leverage the same approach on the massively parallel GPUs or any other shared memory parallel architecture, reducing the run time even further. [Display omitted] • Highly scalable and parallel. • Fast preprocessing and realtime capable. • Usable for either CPU and GPU. [ABSTRACT FROM AUTHOR]

Implicit curve and surface reconstruction has wide applications in computer graphics, which attracts great attention of many researchers. In this paper, we propose the relaxed implicit randomized algebraic reconstruction technique (RIR-ART) for curve and surface reconstruction. The RIR-ART method generates a sequence of curves and surfaces by adjusting the control coefficients with the relaxation parameters iteratively. It is proved that the sequence of curves and surfaces generated by the RIR-ART method with the suitable relaxation parameters converge to the least-norm results in terms of the mean square error. The RIR-ART method can eliminate the extra zero-level sets effectively during its iterative process, and converge faster than the implicit progressive-iterative approximation method (Hamza et al. 2020). Numerical examples present the efficiency and effectiveness of the proposed method. • The RIR-ART method converges in terms of the mean square error. • The extra zero-level sets are eliminated by the RIR-ART method effectively. • The RIR-ART method converges faster than the I-PIA method for the reconstruction. [ABSTRACT FROM AUTHOR]

Surface registration is a fundamental problem in computer graphics and computer-aided design. The problem consists in finding a deformation from one surface to another that preserves some properties. For instance, in our inverse engineering context, we aim at finding the best, as isometric as possible, map between an input triangular model, and a large point cloud acquired on the actual mechanical part being processed. Existing solutions are not able to handle very large models with a good level of precision. We propose a method which is accurate and fast. Our solution combines an efficient iterative energy minimization scheme on a hierarchical decomposition of the problem geometry. Our experiments show that we obtain a fast and efficient algorithm compared to the state-of-the art method, while keeping its numerical accuracy. [Display omitted] • We propose a hierarchical non-rigid isometric registration method. • The registration is performed from surface mesh to point cloud. • The hierarchical optimization makes the method efficient on large datasets. • Meshes and point clouds with millions of vertices are registered in very few minutes. • The method is robust to noisy point cloud with good normal estimators. • Computing the hierarchy is independent of the point cloud and is performed only once. [ABSTRACT FROM AUTHOR]

Modelling of material appearance from reflectance measurements has become increasingly prevalent due to the development of novel methodologies in Computer Graphics. In the last few years, some advances have been made in measuring the light-material interactions, by employing goniometers/reflectometers under specific laboratory's constraints. A wide range of applications benefit from data-driven appearance modelling techniques and material databases to create photorealistic scenarios and physically based simulations. However, important limitations arise from the current material scanning process, mostly related to the high diversity of existing materials in the real-world, the tedious process for material scanning and the spectral characterisation behaviour. Consequently, new approaches are required both for the automatic material acquisition process and for the generation of measured material databases. In this study, a novel approach for material appearance acquisition using hyperspectral data is proposed. A dense 3D point cloud filled with spectral data was generated from the images obtained by an unmanned aerial vehicle (UAV) equipped with an RGB camera and a hyperspectral sensor. The observed hyperspectral signatures were used to recognise natural and artificial materials in the 3D point cloud according to spectral similarity. Then, a parametrisation of Bidirectional Reflectance Distribution Function (BRDF) was carried out by sampling the BRDF space for each material. Consequently, each material is characterised by multiple samples with different incoming and outgoing angles. Finally, an analysis of BRDF sample completeness is performed considering four sunlight positions and 16x16 resolution for each material. The results demonstrated the capability of the used technology and the effectiveness of our method to be used in applications such as spectral rendering and real-word material acquisition and classification. [Display omitted] • A robust pipeline to acquire BRDF samples of real-world materials. • UAV-RGB and hyperspectral sensors are used to observe material-light interactions. • An efficient methodology to capture measured reflectance data in outdoor scenarios. • The coverage of the BRDF space and an estimation for further flights are presented. [ABSTRACT FROM AUTHOR]

Acceleration techniques for Rendering in general, and Ray-Casting in particular, have been the subject of much research in Computer Graphics. Most efforts have been focused on new data structures for efficient ray/scene traversal and intersection. In this paper, we propose an acceleration technique that approximates rendering and that is built around a new feature-based clustering approach. The technique starts preprocessing the scene by grouping elements according to their features using a set of channels based on an information theory-based approach. Then, at run-time, a rendering strategy uses that clustering information to reconstruct the final image, by deciding which areas could take advantage of the coherence in the features and thus, could be interpolated; and which areas require more involved calculations. This process starts with a low-resolution render that is iteratively refined up to the desired resolution by reusing previously computed pixels. Our experimental results show a significant speedup of an order of magnitude, depending on the complexity of the per-pixel calculations, the screen size of the objects, and the number of clusters. Rendering quality and speed directly depend on the number of clusters and the number of steps performed during the reconstruction procedure, and both can easily be set by the user. Our findings show that feature-based clustering can significantly impact rendering speed if samples are chosen to enable interpolation of smooth regions. Our technique, thus, accelerates a range of popular and costly techniques, ranging from texture mapping up to complex ambient occlusion, soft and hard shadow calculations, and it can even be used in conjunction with more traditional acceleration methods. [Display omitted] • Mesh-clustering based Information Theory tools to exploit feature similarity. • User-definable parameters, decoupling them from the actual rendering process. • Multi-pass rendering based on reuse and interpolation of previous results. • Controlling mechanism to guarantee Ray-casting as a lower bound to rendering speed. • Our technique can accommodate both static and animated scenes as well. [ABSTRACT FROM AUTHOR]

The representation of consistent mixed reality (XR) environments requires adequate real and virtual illumination composition in real-time. Estimating the lighting of a real scenario is still a challenge. Due to the ill-posed nature of the problem, classical inverse-rendering techniques tackle the problem for simple lighting setups. However, those assumptions do not satisfy the current state-of-art in computer graphics and XR applications. While many recent works solve the problem using machine learning techniques to estimate the environment light and scene's materials, most of them are limited to geometry or previous knowledge. This paper presents a CNN-based model to estimate complex lighting for mixed reality environments with no previous information about the scene. We model the environment illumination using a set of spherical harmonics (SH) environment lighting, capable of efficiently represent area lighting. We propose a new CNN architecture that inputs an RGB image and recognizes, in real-time, the environment lighting. Unlike previous CNN-based lighting estimation methods, we propose using a highly optimized deep neural network architecture, with a reduced number of parameters, that can learn high complex lighting scenarios from real-world high-dynamic-range (HDR) environment images. We show in the experiments that the CNN architecture can predict the environment lighting with an average mean squared error (MSE) of 7.85 × 10−4 when comparing SH lighting coefficients. We validate our model in a variety of mixed reality scenarios. Furthermore, we present qualitative results comparing relights of real-world scenes. [Display omitted] • Automatic end-to-end method to estimate the environment lighting in XR applications. • A CNN architecture that learns a latent-space of the environment lighting. • A methodology to generate egocentric mixed-reality-views from HDR panoramas. • Real-time lighting estimation that does not make assumptions about the XR scene. [ABSTRACT FROM AUTHOR]

This work presents a method for conforming fluid flow to tensor fields in particle-in-cell simulations. Our focus is on the use of tensor fields as a medium to deflect fluids for computer graphics applications. The anisotropic fluid simulation here presented adapts the Fluid-Implicit-Particle (FLIP) method, by including the tensor field information in the advection and projection steps. Numerical solutions are proposed for adapting the anisotropic advection to the particle-in-cell model and for solving the anisotropic Helmholtz projection for Marker-and-Cell (MAC) grids. Suitable boundary conditions are also provided to ensure proper anisotropic fluid behavior. To evidence the capabilities of this approach, we present simulations with different tensor fields composed of isotropic, linear, and planar tensors. Experimental results show that our method is capable of deflecting or withholding flow in specific directions, according to the characteristics of the underlying tensor field, producing visual outcomes typically encountered in anisotropic media, such as permeable-like effects. [Display omitted] • We present a method for fluid simulation based on tensor fields and the FLIP method. • We give a projection for the MAC grid from the anisotropic Helmholtz decomposition. • We adapt an anisotropic advection to take advantage of particle-in-cell methods. • Tensor-based boundary conditions result in permeable surface effects. • Our method deflects and withholds fluid flow over directions from a tensor field. [ABSTRACT FROM AUTHOR]

• The proposed solution can reconstruct indoor scenes using only a single frame and an RGB camera. • An indoor scene composed by five objects can be reconstructed in just 21 seconds. • The proposed solution obtains a maximum error of 18% for the scaling factor, less than 9cm for the position and less than 18 ∘ for the rotation. • The innovative methodology can be employed for VR and AR applications, thus bridging the gap between the real and virtual words. [Display omitted] Indoor environment reconstruction is a challenging task in Computer Vision and Computer Graphics, especially when Extended Reality (XR) technologies are considered. Current solutions that employ dedicated depth sensors require scanning of the environment and tend to suffer from low resolution and noise, whereas solutions that rely on a single photo of a scene cannot predict the actual position and scale of objects due to scale ambiguity. The proposed system addresses these limitations by allowing the user to capture single views of objects using an Android smartphone equipped with a single RGB camera and supported by Google ARCore. The system includes 1) an Android app tracking the smartphone's position relative to the world, capturing a single RGB image for each object and estimating depth information of the scene, 2) a program running on a server that classifies the framed objects, retrieves the corresponding 3D models from a database and estimates their position, vertical rotation, and scale factor without deforming the shape. The system has been assessed measuring the translational, rotational and scaling errors of the considered objects with respect to the physical ones acting as a ground truth. The main outcomes show that the proposed solution obtains a maximum error of 18% for the scaling factor, less than nine centimeters for the position and less than 18 ∘ for the rotation. These results suggest that the proposed system can be employed for XR applications, thus bridging the gap between the real and virtual worlds. [ABSTRACT FROM AUTHOR]

[Display omitted] Motion Blur is an important effect of photo-realistic rendering. Distribution ray tracing can simulate motion blur very well by integrating light, both over the spatial and the temporal domain. However, increasing the problem by the temporal dimension entails many challenges, particularly in cinematic multi-bounce path tracing of complex scenes, where heavy-weight geometry with complex lighting and even offscreen elements contribute to the final image. In this paper, we propose the Motion DAG (Directed Acyclic Graph), a novel data structure that filters an entire animation sequence of an object, both in the spatial and temporal domain. These Motion DAGs interleave a temporal interval binary tree for filtering time consecutive data and a sparse voxel octree (SVO), which simplifies spatially nearby data. Motion DAGs are generated in a pre-process and can be easily integrated in a conventional physically based path tracer. Our technique is designed to target motion blur of small objects, where coarse representations are sufficient. Specifically, in this scenario our results show that it is possible to significantly reduce both, memory consumption and render time. [ABSTRACT FROM AUTHOR]

• Multi-patch elastic geodesic grids can be used to approximate complicated surfaces. • Multi-patch elastic geodesic grids enable smooth coverage of curved surfaces. • Multi-patch elastic geodesic grids can be deployed sequentially. [Display omitted] Elastic geodesic grids (EGG) are lightweight structures that can be deployed to approximate designer-provided free-form surfaces. Initially, the grids are perfectly flat, during deployment, a curved shape emerges, as grid elements bend and twist. Their layout is based on networks of geodesic curves and is found geometrically. Encoded in the planar grids is the intrinsic shape of the design surface. Such structures may serve purposes like free-form sub-structures, panels, sun and rain protectors, pavilions, etc. However, so far the EGG have only been investigated using a generic set of design surfaces and small-scale desktop models. Some limitations become apparent when considering more sophisticated design surfaces, like from free-form architecture. Due to characteristics like high local curvature or non-geodesic boundaries, they may be captured only poorly by a single EGG. We show how decomposing such surfaces into smaller patches serves as an effective strategy to tackle these problems. We furthermore show that elastic geodesic grids are in fact well suited for this approach. Finally, we present a showcase model of some meters in size and discuss practical aspects concerning fabrication, size, and easy deployment. [ABSTRACT FROM AUTHOR]

• A sufficient and necessary condition for the injectivity of toric volumes is proposed. • The sign of mixed product of three vectors is used to check the compatibility between the sets of lattice points and control points. • An algorithm is proposed to check the injectivity of toric volumes for arbitrary positive weights. • The algorithm is improved based on the property of clean and empty tetrahedron. [Display omitted] Parameterizations, which map parametric domains into certain domains, are widely used in computer aided design, computer aided geometric design, computer graphics, isogeometric analysis, and related fields. The parameterizations of curves, surfaces, and volumes are injective means that they do not have self-intersections. A 3D toric volume is defined via a set of 3D control points with weights that correspond to a set of finite 3D lattice points. Rational tensor product or tetrahedral Bézier volumes are special cases of toric volumes. In this paper, we proved that a toric volume is injective for any positive weights if and only if the lattice points set and control points set are compatible. An algorithm is also presented for checking the compatibility of the two sets by the mixed product of three vectors. Some examples illustrate the effectiveness of the proposed method. Moreover, we improve the algorithm based on the properties and results of clean and empty tetrahedrons in combinatorics. [ABSTRACT FROM AUTHOR]

3D shape generation, vital in fields including computer graphics, industrial design, and robotics, has seen a significant growth due to deep learning advancements. Nevertheless, a prevailing challenge in this area lies in its heavy reliance on extensive data for training. Consequently, the ability to generate 3D shapes with a limited quantity of training samples emerges as a desirable objective. The aim of this research is to design deep generative models capable of learning from a single reference 3D shape, thereby eliminating the requirement for sizeable datasets. Drawing inspiration from contemporary Generative Adversarial Networks (GANs) that operate on individual 3D shapes in a coarse-to-fine manner hierarchically, we propose a novel wavelet-based framework for single 3D shape generation, which preserves the global shape structure whilst inducing local variability. Our key observation is that, through wavelet decomposition, the low-frequency components of two inputs, where one input is a corrupted version of the other, are very similar. This similarity enables reconstruction of the uncorrupted input by leveraging the low-frequency components of the corrupted version. This observation motivates us to propose the wavelet decomposition of the 2D tri-plane feature maps of a given 3D shape, followed by the synthesis of new tri-plane feature maps for shape generation. To the best of our knowledge, this work represents the first endeavor to incorporate wavelet analysis into a deep generative model for the purpose of generating novel 3D shapes with a single example. Furthermore, we adapt data augmentation and Coulomb adversarial generative loss to facilitate training and generation procedures. We demonstrate the effectiveness of our approach by generating diverse 3D shapes and conducting quantitative comparisons with established baseline methods. Our implementation is available at https://github.com/hhuang-code/SinWavelet. [Display omitted] • We propose a wavelet generative model to generate novel shapes with a single input. • We adapt data augmentation and Coulomb adversarial loss to facilitate training. • Experiments show that our model can generate diverse 3D shapes in various categories. [ABSTRACT FROM AUTHOR]

• Computer graphics. • Computer vision. • Deep learning. • 3D reconstruction. • Virtual human digitization. • Generative adversarial network. [Display omitted] With the development of AR/VR technologies, a reliable and straightforward way to digitize a three-dimensional human body is in high demand. Most existing methods use complex equipment and sophisticated algorithms, but this is impractical for everyday users. In this paper, we propose a pipeline that reconstructs a 3D human shape avatar from a single image. Our approach simultaneously reconstructs the three-dimensional human geometry and whole body texture map with only a single RGB image as input. We first segment the human body parts from the image and then obtain an initial body geometry by fitting the segment to a parametric model. Next, we warp the initial geometry to the final shape by utilizing a silhouette-based dense correspondence. Finally, to infer invisible back texture from a frontal image, we propose a network called InferGAN. Based on human semantic information, we also propose a method to handle partial occlusion by reconstructing the occluded body parts separately. Comprehensive experiments demonstrate that our solution is robust and effective on both public and our own datasets. Our human avatars can be easily rigged and animated using MoCap data. We have developed a mobile application that demonstrates this capability for AR applications. [ABSTRACT FROM AUTHOR]

[Display omitted] • Reinforcement Learning for traffic simulations. • Hybrid deep learning and optimization method for 3D human pose and shape reconstruction in simulated depth images. [ABSTRACT FROM AUTHOR]

• Propose an immersive method of sketch-based tree modeling in virtual reality. • Present a series of algorithms to deal with the problems of generating branch geometry for real-time 3D sketching. • Provide a bidirectional ray-hit algorithm to determine the branch radius in real-time. • Present a twigs generation algorithm to randomly generate twigs based on Perlin noise and parent branch direction. Tree modeling has been a widely-discussed topic in computer graphics. However, with existing methods, the modeling process is occupied with complex data collection and tedious parameter adjustment, lacking a rich sensory modeling experience. In this paper, we propose an approach to sketch-based tree modeling in an immersive virtual reality environment, aiming to lower the difficulty of modeling and enhance the immersion in the designing process. We first present a sketch sampling and points optimization algorithm to obtain the skeleton of the branch in 3D space. As generating geometry along the skeleton, we apply a vector-projection method to fix the branch polygon twisting. Then, we introduce a bidirectional ray-hit algorithm to determine the branch radius as real-time sketching. To generate random twigs on branches, we introduce a twigs generation algorithm based on Perlin noise and the parent branch direction. Finally, we design a series of interactive methods for users to create tree models in a 3D virtual scene with the VR HMD and controller. Experimental results indicate that our approach can accomplish creating realistic tree models in real-time. The interactive and immersive modeling experience enables users to readily convey their ideas on tree structures in a simple and direct way of sketching. Visualizing and modifying the real-time generated branch results can contribute to provoking the inspiration for creation. [ABSTRACT FROM AUTHOR]

• Use non-realistic computer graphics to generate training samples for object detection. • Investigate the impact of context when training deep models with synthetic samples. • Experiments are performed in several well-known traffic light datasets. • Our approach achieves results comparable to those that use real-world training data. Deep neural networks come as an effective solution to many problems associated with autonomous driving. By providing real image samples with traffic context to the network, the model learns to detect and classify elements of interest, such as pedestrians, traffic signs, and traffic lights. However, acquiring and annotating real data can be extremely costly in terms of time and effort. In this context, we propose a method to generate artificial traffic-related training data for deep traffic light detectors. This data is generated using basic non-realistic computer graphics to blend fake traffic scenes on top of arbitrary image backgrounds that are not related to the traffic domain. Thus, a large amount of training data can be generated without annotation efforts. Furthermore, it also tackles the intrinsic data imbalance problem in traffic light datasets, caused mainly by the low amount of samples of the yellow state. Experiments show that it is possible to achieve results comparable to those obtained with real training data from the problem domain, yielding an average mAP and an average F1-score which are each nearly 4 p.p. higher than the respective metrics obtained with a real-world reference model. [ABSTRACT FROM AUTHOR]

• Introduction of Riemannian global illumination. • A path tracer algorithm to approximate the solution of the Riemannian global illumination. • A map of the algorithm to the GPU, using the RTX platform. • We apply the Non-Euclidean path tracer to render "Photorealistic" inside views of the 3D flat torus, Poincaré sphere, and the hyperbolic mirrored dodecahedron. These are examples of the Euclidean, spherical, and hyperbolic spaces: the Thurston classical geometries. This paper presents a novel path tracer algorithm for immersive visualization of Riemannian manifolds. To do this, we introduce Riemannian illumination, a generalization of classical Computer Graphics illumination models. In this context, global light transport is expressed by extending the rendering equation to Riemannian manifolds. Using Monte Carlo integration to solve this equation results in the novel path tracer for Non-Euclidean spaces. We discuss its basic principles, as well as the general CPU algorithm. Additionally, we discuss in detail how to implement a GPU version, using the RTX pipeline. Finally, we apply the algorithm to render "photorealistic" inside views of the flat torus, Poincaré sphere, and the hyperbolic mirrored dodecahedron. These are examples of Euclidean, spherical, and hyperbolic spaces: the Thurston classical geometries. [ABSTRACT FROM AUTHOR]

• Anatomical muscle models are important for facial simulation and animation. • Parametrizing the muscle activations remains a challenge. • Solving for the blending geometry of muscles reduces the space of activations. • Blending geometry information can be reused to fit new expression scans. Modeling of the human face is a challenging yet important problem in computer graphics. Building accurate muscle models for physics-based simulation of the face is a problem that either requires a lot of manual effort or drastic over-parameterization of the muscles to achieve desirable results. In this work, we reduce the number of parameters required to build personalized muscle models by taking into account the blending of the fine muscles and passive tissue when we solve for the muscle activations. We begin by adapting an anatomical template model to a neutral scan of a subject. Then, we solve an inverse physics problem using several scans simultaneously to solve for both the muscle activations and the geometry matrix representing blending of the muscles. Finally, we demonstrate that this geometry matrix can be used on new, previously unseen scans to solve for only the muscle activations. This greatly reduces the number of parameters that must be solved for compared to previous works while requiring no additional manual effort in constructing the muscles. [ABSTRACT FROM AUTHOR]

• Introduction of Riemannian shading, generalizing classic Computer Graphics concepts. • Proposal of a ray tracing model for immersive visualization of Riemannian manifolds. • Realtime VR visualization of Nil, Sol, and SL2: the most non-trivial Thurston geometries. In this work, we propose a novel ray tracing model for immersive visualization of Riemannian manifolds. To do this we introduce Riemannian ray tracing, a generalization of the classic Computer Graphics concept. Specifically, our model is capable of interactive real-time VR visualizations of Nil, Sol , and S L 2 (R) ˜ , Thurston's most nontrivial geometries. These experiences have the potential to allow insights with impact in physics/cosmology research, education, special effects, and games, among other aspects. The Riemannian ray tracing is implemented using the ray-tracing capabilities of the NVidia RTX platform. We discuss the general algorithm in CPU and show how to map the computations to the RTX pipeline. [ABSTRACT FROM AUTHOR]