EG 2020 - Short Papers
Permanent URI for this collection
Browse
Browsing EG 2020 - Short Papers by Issue Date
Now showing 1 - 20 of 25
Results Per Page
Sort Options
Item Adversarial Generation of Continuous Implicit Shape Representations(The Eurographics Association, 2020) Kleineberg, Marian; Fey, Matthias; Weichert, Frank; Wilkie, Alexander and Banterle, FrancescoThis work presents a generative adversarial architecture for generating three-dimensional shapes based on signed distance representations. While the deep generation of shapes has been mostly tackled by voxel and surface point cloud approaches, our generator learns to approximate the signed distance for any point in space given prior latent information. Although structurally similar to generative point cloud approaches, this formulation can be evaluated with arbitrary point density during inference, leading to fine-grained details in generated outputs. Furthermore, we study the effects of using either progressively growing voxel- or point-processing networks as discriminators, and propose a refinement scheme to strengthen the generator's capabilities in modeling the zero iso-surface decision boundary of shapes. We train our approach on the SHAPENET benchmark dataset and validate, both quantitatively and qualitatively, its performance in generating realistic 3D shapes.Item Learning Body Shape and Pose from Dense Correspondences(The Eurographics Association, 2020) Yoshiyasu, Yusuke; Gamez, Lucas; Wilkie, Alexander and Banterle, FrancescoIn this paper, we address the problem of learning 3D human pose and body shape from 2D image dataset, without having to use 3D supervisions (body shape and pose) which are in practice difficult to obtain. The idea is to use dense correspondences between image points and a body surface, which can be annotated on in-the-wild 2D images, to extract, aggregate and learn 3D information such as body shape and pose from them. To do so, we propose a training strategy called "deform-and-learn" where we alternate deformable surface registration and training of deep convolutional neural networks (ConvNets). Experimental results showed that our method is comparable to previous semi-supervised techniques that use 3D supervision.Item Organic Narrative Charts(The Eurographics Association, 2020) Bolte, Fabian; Bruckner, Stefan; Wilkie, Alexander and Banterle, FrancescoStoryline visualizations display the interactions of groups and entities and their development over time. Existing approaches have successfully adopted the general layout from hand-drawn illustrations to automatically create similar depictions. Ward Shelley is the author of several diagrammatic paintings that show the timeline of art-related subjects, such as Downtown Body, a history of art scenes. His drawings include many stylistic elements that are not covered by existing storyline visualizations, like links between entities, splits and merges of streams, and tags or labels to describe the individual elements. We present a visualization method that provides a visual mapping for the complex relationships in the data, creates a layout for their display, and adopts a similar styling of elements to imitate the artistic appeal of such illustrations.We compare our results to the original drawings and provide an open-source authoring tool prototype.Item Deep-Eyes: Fully Automatic Anime Character Colorization with Painting of Details on Empty Pupils(The Eurographics Association, 2020) Akita, Kenta; Morimoto, Yuki; Tsuruno, Reiji; Wilkie, Alexander and Banterle, FrancescoMany studies have recently applied deep learning to the automatic colorization of line drawings. However, it is difficult to paint empty pupils using existing methods because the networks are trained with pupils that have edges, which are generated from color images using image processing. Most actual line drawings have empty pupils that artists must paint in. In this paper, we propose a novel network model that transfers the pupil details in a reference color image to input line drawings with empty pupils. We also propose a method for accurately and automatically coloring eyes. In this method, eye patches are extracted from a reference color image and automatically added to an input line drawing as color hints using our eye position estimation network.Item EUROGRAPHICS 2020: Short Papers Frontmatter(Eurographics Association, 2020) Wilkie, Alexander; Banterle, Francesco; Wilkie, Alexander and Banterle, FrancescoItem Pair Correlation Functions with Free-Form Boundaries for Distribution Inpainting and Decomposition(The Eurographics Association, 2020) Nicolet, Baptiste; Ecormier-Nocca, Pierre; Memari, Pooran; Cani, Marie-Paule; Wilkie, Alexander and Banterle, FrancescoPair Correlation Functions (PCF) have been recently spreading as a reliable representation for distributions, enabling the efficient synthesis of point-sets, vector textures and object placement from examples. In this work we introduce a triangulationbased local filtering method to extend PCF-based analysis to exemplars with free-form boundaries. This makes PCF applicable to new problems such as the inpainting of missing parts in an input distribution, or the decomposition of complex, non-homogeneous distributions into a set of coherent classes, in which each category of points can be studied together with their intra and inter-class correlations.Item Triplanar Displacement Mapping for Terrain Rendering(The Eurographics Association, 2020) Weiss, Sebastian; Bayer, Florian; Westermann, Rüdiger; Wilkie, Alexander and Banterle, FrancescoHeightmap-based terrain representations are common in computer games and simulations. However, adding geometric details to such a representation during rendering on the GPU is difficult to achieve. In this paper, we propose a combination of triplanar mapping, displacement mapping, and tessellation on the GPU, to create extruded geometry along steep faces of heightmap-based terrain fields on-the-fky during rendering. The method allows rendering geometric details such as overhangs and boulders, without explicit triangulation. We further demonstrate how to handle collisions and shadows for the enriched geometry.Item Interactive Flat Coloring of Minimalist Neat Sketches(The Eurographics Association, 2020) Parakkat, Amal Dev; Madipally, Prudhviraj; Gowtham, Hari Hara; Cani, Marie-Paule; Wilkie, Alexander and Banterle, FrancescoWe introduce a simple Delaunay-triangulation based algorithm for the interactive coloring of neat line-art minimalist sketches, ie. vector sketches that may include open contours. The main objective is to minimize user intervention and make interaction as natural as with the flood-fill algorithm while extending coloring to regions with open contours. In particular, we want to save the user from worrying about parameters such as stroke weight and size. Our solution works in two steps, 1) a segmentation step in which the input sketch is automatically divided into regions based on the underlying Delaunay structure and 2) the interactive grouping of neighboring regions based on user input. More precisely, a region adjacency graph is computed from the segmentation result, and is interactively partitioned based on user input to generate the final colored sketch. Results show that our method is as natural as a bucket fill tool and powerful enough to color minimalist sketches.Item Space-Time Blending for Heterogeneous Objects(The Eurographics Association, 2020) Tereshin, Alexander; Anderson, Eike; Pasko, Alexander; Adzhiev, Valery; Wilkie, Alexander and Banterle, FrancescoSpace-time blending (STB) is an established technique allowing to implement a metamorphosis operation between geometric shapes. In this paper we significantly extend the STB method to make it possible to deal with heterogeneous objects, which are volumetric objects with attributes representing their physical properties. The STB method, used for geometry transformation, is naturally combined with space-time transfinite interpolation, used for attribute (e.g. colour) transformation. Geometry and attribute transformations are interconnected and happen simultaneously in an higher dimensional specific STB space. We use hybrid function representation, unifying function representation with signed distance fields and with adaptively sampled distance fields. We show how the new method works by applying it to 4D animated Cubism.Item Conservative Ray Batching using Geometry Proxies(The Eurographics Association, 2020) Molenaar, Mathijs; Eisemann, Elmar; Wilkie, Alexander and Banterle, FrancescoWe present a method for improving batched ray traversal as was presented by Pharr et al. [PKGH97]. We propose to use conservative proxy geometry to more accurately determine whether a ray has a possibility of hitting any geometry that is stored on disk. This prevents unnecessary disk loads and thus reduces the disk bandwidth.Item First Order Signed Distance Fields(The Eurographics Association, 2020) Bán, Róbert; Valasek, Gábor; Wilkie, Alexander and Banterle, FrancescoThis paper investigates a first order generalization of signed distance fields. We show that we can improve accuracy and storage efficiency by incorporating the spatial derivatives of the signed distance function into the distance field samples. We show that a representation in power basis remains invariant under barycentric combination, as such, it is interpolated exactly by the GPU. Our construction is applicable in any geometric setting where point-surface distances can be queried. To emphasize the practical advantages of this approach, we apply our results to signed distance field generation from triangular meshes. We propose storage optimization approaches and offer a theoretical and empirical accuracy analysis of our proposed distance field type in relation to traditional, zero order distance fields. We show that the proposed representation may offer an order of magnitude improvement in storage while retaining the same precision as a higher resolution distance field.Item UV Completion with Self-referenced Discrimination(The Eurographics Association, 2020) Kang, Jiwoo; Lee, Seongmin; Lee, Sanghoon; Wilkie, Alexander and Banterle, FrancescoA facial UV map is used in many applications such as facial reconstruction, synthesis, recognition, and editing. However, it is difficult to collect a number of the UVs needed for accuracy using 3D scan device, or a multi-view capturing system should be required to construct the UV. An occluded facial UV with holes could be obtained by sampling an image after fitting a 3D facial model by recent alignment methods. In this paper, we introduce a facial UV completion framework to train the deep neural network with a set of incomplete UV textures. By using the fact that the facial texture distributions of the left and right half-sides are almost equal, we devise an adversarial network to model the complete UV distribution of the facial texture. Also, we propose the self-referenced discrimination scheme that uses the facial UV completed from the generator for training real distribution. It is demonstrated that the network can be trained to complete the facial texture with incomplete UVs comparably to when utilizing the ground-truth UVs.Item MEPP2: A Generic Platform for Processing 3D Meshes and Point Clouds(The Eurographics Association, 2020) Vidal, Vincent; Lombardi, Eric; Tola, Martial; Dupont, Florent; Lavoué, Guillaume; Wilkie, Alexander and Banterle, FrancescoIn this paper, we present MEPP2, an open-source C++ software development kit (SDK) for processing and visualizing 3D surface meshes and point clouds. It provides both an application programming interface (API) for creating new processing filters and a graphical user interface (GUI) that facilitates the integration of new filters as plugins. Static and dynamic 3D meshes and point clouds with appearance-related attributes (color, texture information, normal) are supported. The strength of the platform is to be generic programming oriented. It offers an abstraction layer, based on C++ Concepts, that provides interoperability over several third party mesh and point cloud data structures, such as OpenMesh, CGAL, and PCL. Generic code can be run on all data structures implementing the required concepts, which allows for performance and memory footprint comparison. Our platform also permits to create complex processing pipelines gathering idiosyncratic functionalities of the different libraries. We provide examples of such applications. MEPP2 runs on Windows, Linux & Mac OS X and is intended for engineers, researchers, but also students thanks to simple use, facilitated by the proposed architecture and extensive documentation.Item Neural Smoke Stylization with Color Transfer(The Eurographics Association, 2020) Christen, Fabienne; Kim, Byungsoo; Azevedo, Vinicius C.; Solenthaler, Barbara; Wilkie, Alexander and Banterle, FrancescoArtistically controlling fluid simulations requires a large amount of manual work by an artist. The recently presented transportbased neural style transfer approach simplifies workflows as it transfers the style of arbitrary input images onto 3D smoke simulations. However, the method only modifies the shape of the fluid but omits color information. In this work, we therefore extend the previous approach to obtain a complete pipeline for transferring shape and color information onto 2D and 3D smoke simulations with neural networks. Our results demonstrate that our method successfully transfers colored style features consistently in space and time to smoke data for different input textures.Item Frequency-Aware Reconstruction of Fluid Simulations with Generative Networks(The Eurographics Association, 2020) Biland, Simon; Azevedo, Vinicius C.; Kim, Byungsoo; Solenthaler, Barbara; Wilkie, Alexander and Banterle, FrancescoConvolutional neural networks were recently employed to fully reconstruct fluid simulation data from a set of reduced parameters. However, since (de-)convolutions traditionally trained with supervised l1-loss functions do not discriminate between low and high frequencies in the data, the error is not minimized efficiently for higher bands. This directly correlates with the quality of the perceived results, since missing high frequency details are easily noticeable. In this paper, we analyze the reconstruction quality of generative networks and present a frequency-aware loss function that is able to focus on specific bands of the dataset during training time. We show that our approach improves reconstruction quality of fluid simulation data in mid-frequency bands, yielding perceptually better results while requiring comparable training time.Item Procedural 3D Asteroid Surface Detail Synthesis(The Eurographics Association, 2020) Li, Xi-zhi; Weller, René; Zachmann, Gabriel; Wilkie, Alexander and Banterle, FrancescoWe present a novel noise model to procedurally generate volumetric terrain on implicit surfaces. The main idea is to combine a novel Locally Controlled 3D Spot noise (LCSN) for authoring the macro structures and 3D Gabor noise to add micro details. More specifically, a spatially-defined kernel formulation in combination with an impulse distribution enables the LCSN to generate arbitrary size craters and boulders, while the Gabor noise generates stochastic Gaussian details. The corresponding metaball positions in the underlying implicit surface preserve locality to avoid the globality of traditional procedural noise textures, which yields an essential feature that is often missing in procedural texture based terrain generators. Furthermore, different noise-based primitives are integrated through operators, i.e. blending, replacing, or warping into the complex volumetric terrain. The result is a completely implicit representation and, as such, has the advantage of compactness as well as flexible user control. We applied our method to generating high quality asteroid meshes with fine surface details.Item On Learning the Best Local Balancing Strategy(The Eurographics Association, 2020) Murray, David; Benzait, Sofiane; Pacanowski, Romain; Granier, Xavier; Wilkie, Alexander and Banterle, FrancescoFast computation of light propagation using Monte Carlo techniques requires finding the best samples from the space of light paths. For the last 30 years, numerous strategies have been developed to address this problem but choosing the best one is really scene-dependent. Multiple Importance Sampling (MIS) emerges as a potential generic solution by combining different weighted strategies, to take advantage of the best ones. Most recent work have focused on defining the best weighting scheme. Among them, two paper have shown that it is possible, in the context of direct illumination, to estimate the best way to balance the number of samples between two strategies, on a per-pixel basis. In this paper, we extend this previous approach to Global Illumination and to three strategies.Item Interactive Assembly and Animation of 3D Digital Garments(The Eurographics Association, 2020) Nylén, Oskar; Pall, Pontus; Ishiwaka, Yuko; Suda, Kazuto; Fratarcangeli, Marco; Wilkie, Alexander and Banterle, FrancescoWe present a novel real-time tool for sewing together 2D patterns, enabling quick assembly of visually plausible, interactively animated garments for virtual characters. The process is assisted by ad-hoc visual hints and allows designers to import 2D patterns from any CAD-tool, connect them using seams around a 3D character with any body type, and assess the overall quality during the character animation. The cloth is numerically simulated including robust modeling of contact of the cloth with itself and with the character's body. Overall, our tool allows for fast prototyping of virtual garments, achieving immediate feedback on their behaviour and visual quality on an animated character, in effect speeding up the content production pipeline for visual effects applications involving clothed characters.Item Photon Mapping Superluminal Particles(The Eurographics Association, 2020) Waldemarson, Gustaf; Doggett, Michael; Wilkie, Alexander and Banterle, FrancescoOne type of light source that remains largely unexplored in the field of light transport rendering is the light generated by superluminal particles, a phenomenon more commonly known as Cherenkov radiation [Cˇ37]. By re-purposing the Frank-Tamm equation [FT91] for rendering, the energy output of these particles can be estimated and consequently mapped to photons, making it possible to visualize the brilliant blue light characteristic of the effect. In this paper we extend a stochastic progressive photon mapper and simulate the emission of superluminal particles from a source object close to a medium with a high index of refraction. In practice, the source is treated as a new kind of light source, allowing us to efficiently reuse existing photon mapping methods.Item Compression and Real-Time Rendering of Inward Looking Spherical Light Fields(The Eurographics Association, 2020) Hajisharif, Saghi; Miandji, Ehsan; Baravadish, Gabriel; Larsson, Per; Unger, Jonas; Wilkie, Alexander and Banterle, FrancescoPhotorealistic rendering is an essential tool for immersive virtual reality. In this regard, the data structure of choice is typically light fields since they contain multidimensional information about the captured environment that can provide motion parallax and view-dependent information such as highlights. There are various ways to acquire light fields depending on the nature of the scene, limitations on the capturing setup, and the application at hand. Our focus in this paper is on full-parallax imaging of large-scale static objects for photorealistic real-time rendering. To this end, we introduce and simulate a new design for capturing inward-looking spherical light fields, and propose a system for efficient compression and real-time rendering of such data using consumer-level hardware suitable for virtual reality applications.