EG2016

Permanent URI for this community

https://diglib.eg.org/handle/10.2312/15062

Browse

Now showing 1 - 20 of 116

A Trip to Arts for Computer Graphics Students
(The Eurographics Association, 2016) Svobodova, Lucie; Slavik, Pavel; Zara, Jiri; Beatriz Sousa Santos and Jean-Michel Dischler
Complex and sophisticated projects, on which information technologists and artists meet and cooperate, often require an understanding of possible problems and solutions seen from both sides. While artists (creators) are usually somewhat familiar with current technologies, e.g. the Computer Graphics (CG) field, programmers are not well educated in the arts and their understanding of artistic needs is thus insufficient. In order to improve the education of CG students towards fine arts, we have created a BSc course on Art and Graphics design and have been running it for seven years. We consider this approach to education at a technologically oriented university as unusual and rare, but highly valuable for CG students. This paper describes the content of the course, summarizes the experience gained, and evaluates its usefulness for CG students in the subsequent MSc study program.
Sketch-based Modeling
(The Eurographics Association, 2016) Cordier, Frederic; Singh, Karan; Etem, Even; Cani, Marie-Paule; Gingold, Yotam; Augusto Sousa and Kadi Bouatouch
Sketching is one of the most natural ways to exchange ideas. It has been used by human beings since prehistory. Research has shown that human beings have an inherent ability to understand sketches. This is why sketch-based interfaces for 3D modeling are so appealing; creating and animating 3D shapes could become as simple as drawing with a tablet and a digital pen. The purpose of this tutorial is to explore the most important aspects of sketch-based modeling, from the preprocessing of sketch strokes to the problem of 3D reconstruction. We will first explain some aspects of how humans interpret sketches. The second part of the tutorial will be dedicated to the problem of filtering and processing strokes. Other parts of the tutorial will focus on the sketch-based modeling of curves and surfaces using multi-view and single-view sketches. Sketch-based modeling using prior-knowledge will be also discussed; this class of methods is particularly well adapted to the 3D reconstruction of complex shapes. The last part of the tutorial addresses sketch-based interfaces for editing 3D shapes.
EUROGRAPHICS 2016: Posters Frontmatter
(Eurographics Association, 2016) Luis Gonzaga Magalhães; Rafał Mantiuk;
Space-Time Co-Segmentation of Articulated Point Cloud Sequences
(The Eurographics Association and John Wiley & Sons Ltd., 2016) Yuan, Qing; Li, Guiqing; Xu, Kai; Chen, Xudong; Huang, Hui; Joaquim Jorge and Ming Lin
Consistent segmentation is to the center of many applications based on dynamic geometric data. Directly segmenting a raw 3D point cloud sequence is a challenging task due to the low data quality and large inter-frame variation across the whole sequence. We propose a local-to-global approach to co-segment point cloud sequences of articulated objects into near-rigid moving parts. Our method starts from a per-frame point clustering, derived from a robust voting-based trajectory analysis. The local segments are then progressively propagated to the neighboring frames with a cut propagation operation, and further merged through all frames using a novel space-time segment grouping technqiue, leading to a globally consistent and compact segmentation of the entire articulated point cloud sequence. Such progressive propagating and merging, in both space and time dimensions, makes our co-segmentation algorithm especially robust in handling noise, occlusions and pose/view variations that are usually associated with raw scan data.
Skyglow: Towards a Night-time Illumination Model for Urban Environments
(The Eurographics Association, 2016) Minor, Tom; Poncelet, Robert R.; Anderson, Eike Falk; Luis Gonzaga Magalhaes and Rafal Mantiuk
For night-time scenes in computer graphics there exist few consistent models or implementations for sky illumination, and those that do exist lack the feature of light pollution from artificial light sources. We present initial results for a physically-based night sky model including this ''skyglow''. Our model extends the existing models with the aforementioned ''skyglow'' from artificial light sources, using a technique derived from equations developed in the field of astronomy and adapted for a computer graphics context. Our current model has been implemented for Pixar's RenderMan renderer and also been trialled with ShaderToy.
Drift-Diffusion Based Real-Time Dynamic Terrain Deformation
(The Eurographics Association, 2016) Gilardi, Marco; Watten, Phil L.; Newbury, Paul; T. Bashford-Rogers and L. P. Santos
In the natural world, terrains are dynamic entities which change their morphology due to their interaction with other agents in the environment. However, in real-time applications terrains are often represented as static meshes, which present no interaction capabilities. This paper presents a novel real-time 2D method for dynamic terrain simulations, aimed for applications in the entertainment industry. This method is based on a Dynamically-Displaced Height-map and on the numerical solutions, obtained using an Euler method, of a modified drift-diffusion equation. The method allows objects to interact with the terrain and to deform it in real time, it is easy to implement and generates different kinds of realistic tracks depending on the soil composition.
EUROGRAPHICS 2016: State of the Art Reports Frontmatter
(The Eurographics Association and John Wiley & Sons Ltd., 2016) Joaquim Madeira; Gustavo Patow;
A 3D Morphable Model of the Eye Region
(The Eurographics Association, 2016) Wood, Erroll; Baltrušaitis, Tadas; Morency, Louis-Philippe; Robinson, Peter; Bulling, Andreas; Luis Gonzaga Magalhaes and Rafal Mantiuk
We present the first 3D morphable model that includes the eyes, enabling gaze estimation and gaze re-targetting from a single image. Morphable face models are a powerful tool and are used for a range of tasks including avatar animation and facial expression transfer. However, previous work has avoided the eyes, even though they play an important role in human communication. We built a new morphable model of the facial eye-region from high-quality head scan data, and combined this with a parametric eyeball model constructed from anatomical measurements and iris photos. We fit our models to an input RGB image, solving for shape, texture, pose, and scene illumination simultaneously. This provides us with an estimate of where a person is looking in a 3D scene without per-user calibration - a still unsolved problem in computer vision. It also allows us to re-render a person's eyes with different parameters, thus redirecting their perceived attention.
Peer Review: Does it really help students?
(The Eurographics Association, 2016) Kenwright, Ben; Beatriz Sousa Santos and Jean-Michel Dischler
Student peer review has long been a method for increasing student engagement and work quality.We present notes on teaching tips and techniques using peer review as a means way to engage students interest in the area of computer graphics and interactive animation. We address questions, such as, when feedback fails, why students should be ‘trained’ on feedback, and what constitutes a ‘constructive’ review. We present a case study around the structure and workings of a module - and its success in encouraging collaborative working, group discussions, public engagement (e.g., through wikis and events), and peer review work.
Tonal Art Maps with Image Space Strokes
(The Eurographics Association, 2016) Szécsi, László; Szirányi, Marcell; Kacsó, Ágota; Luis Gonzaga Magalhaes and Rafal Mantiuk
This paper presents a hybrid hatching solution that uses robust and fast texture space hatching to gather stroke fragments, but fits stylized brush strokes over those fragments in image space. Thus we obtain a real-time solution that avoids the challenges associated with hidden stroke removal in image space approaches, but allows for the artistic stylization of strokes exceeding the limitations of texture space methods. This includes strokes running over outlines or behind occluders, uniquely random strokes, and adherence to image space brush properties.
Subjective and Objective Evaluation of Multi-exposure High Dynamic Range Image Deghosting Methods
(The Eurographics Association, 2016) Karaduzovic-Hadziabdic, Kanita; Telalovic, Jasminka Hasic; Mantiuk, Rafal; T. Bashford-Rogers and L. P. Santos
To avoid motion artefacts when merging multiple exposures into an HDR image, a number of deghosting algorithms have been proposed. These algorithms, however, do not work equally well on all types of scenes, and some may even introduce additional artefacts. Even though subjective methods of evaluation provide reliable means of testing, they need to be repeated for each new proposed method or even its slight modification and are cumbersome to perform. In this work, we evaluate several computational approaches of quantitative evaluation of multi-exposure HDR deghosting algorithms and demonstrate their results on five state-of-the-art algorithms.The quality of HDR images produced by deghosting methods is measured in a subjective experiment, and then evaluated using five objective metrics. The most reliable metrics is then selected by testing correlation between subjective and objective metric scores.
Efficient Point based Global Illumination on Intel MIC Architecture
(The Eurographics Association, 2016) Xu, Xiang; Wang, Pei; Wang, Beibei; Wang, Lu; Tu, Changhe; Meng, Xiangxu; Boubekeur, Tamy; Luis Gonzaga Magalhaes and Rafal Mantiuk
Point-Based Global Illumination (PBGI) is a popular rendering method in special effects and motion picture productions. The tree-cut computation is in general the most time consuming part of this algorithm, but it can be formulated for efficient parallel execution, in particular regarding wide-SIMD hardware. In this context, we propose several vectorization schemes, namely single, packet and hybrid, to maximize the utilization of modern CPU architectures. While for the single scheme, 16 nodes from the hierarchy are processed for a single receiver in parallel, the packet scheme handles one node for 16 receivers. These two schemes work well for scenes having smooth geometry and diffuse material. When the scene contains high frequency bumps maps and glossy reflections, we use a hybrid vectorization method. We conduct experiments on an Intel Many Integrated Core architecture and report preliminary results on several scenes, showing that up to a 3x speedup can be achieved when compared with non-vectorized execution.
Learning 3D Deformation of Animals from 2D Images
(The Eurographics Association and John Wiley & Sons Ltd., 2016) Kanazawa, Angjoo; Kovalsky, Shahar; Basri, Ronen; Jacobs, David; Joaquim Jorge and Ming Lin
Understanding how an animal can deform and articulate is essential for a realistic modification of its 3D model. In this paper, we show that such information can be learned from user-clicked 2D images and a template 3D model of the target animal. We present a volumetric deformation framework that produces a set of new 3D models by deforming a template 3D model according to a set of user-clicked images. Our framework is based on a novel locally-bounded deformation energy, where every local region has its own stiffness value that bounds how much distortion is allowed at that location. We jointly learn the local stiffness bounds as we deform the template 3D mesh to match each user-clicked image. We show that this seemingly complex task can be solved as a sequence of convex optimization problems. We demonstrate the effectiveness of our approach on cats and horses, which are highly deformable and articulated animals. Our framework produces new 3D models of animals that are significantly more plausible than methods without learned stiffness.
Anisotropic Diffusion Descriptors
(The Eurographics Association and John Wiley & Sons Ltd., 2016) Boscaini, Davide; Masci, Jonathan; Rodolà, Emanuele; Bronstein, Michael M.; Cremers, Daniel; Joaquim Jorge and Ming Lin
Spectral methods have recently gained popularity in many domains of computer graphics and geometry processing, especially shape processing, computation of shape descriptors, distances, and correspondence. Spectral geometric structures are intrinsic and thus invariant to isometric deformations, are efficiently computed, and can be constructed on shapes in different representations. A notable drawback of these constructions, however, is that they are isotropic, i.e., insensitive to direction. In this paper, we show how to construct direction-sensitive spectral feature descriptors using anisotropic diffusion on meshes and point clouds. The core of our construction are directed local kernels acting similarly to steerable filters, which are learned in a task-specific manner. Remarkably, while being intrinsic, our descriptors allow to disambiguate reflection symmetries. We show the application of anisotropic descriptors for problems of shape correspondence on meshes and point clouds, achieving results significantly better than state-of-the-art methods.
Garment Transfer for Quadruped Characters
(The Eurographics Association, 2016) Narita, Fumiya; Saito, Shunsuke; Kato, Takuya; Fukusato, Tsukasa; Morishima, Shigeo; T. Bashford-Rogers and L. P. Santos
Modeling clothing to characters is one of the most time-consuming tasks for artists in 3DCG animation production. Transferring existing clothing models is a simple and powerful solution to reduce labor. In this paper, we propose a method to generate a clothing model for various characters from a single template model. Our framework consists of three steps: scale measurement, clothing transformation, and texture preservation. By introducing a novel measurement of the scale deviation between two characters with different shapes and poses, our framework achieves pose-independent transfer of clothing even for quadrupeds (e.g., from human to horse). In addition to a plausible clothing transformation method based on the scale measurement, our method minimizes texture distortion resulting from large deformation. We demonstrate that our system is robust for a wide range of body shapes and poses, which is challenging for current state-of-the-art methods.
Trajectory Data Visualization on Mobile Devices with Animated Maps
(The Eurographics Association, 2016) Gonçalves, Tiago; Afonso, Ana Paula; Ferreira, António; Vieira, Ana Rita; T. Bashford-Rogers and L. P. Santos
With the increasing popularity of mobile devices (like smartphones and tablets) and georeferenced applications, more people record and analyse their own movement data. This pattern is noticeable with the increasing usage of mobile applications that, in addition to record the evolution of a person's location over time, also allow the visualization of that information, typically, in the form of 2D static maps, complemented with various representations to extract knowledge from the data. Despite the various studies addressing spatio-temporal data visualization, its application on mobile devices for the representation of personal trajectory data is still somewhat unexplored. Animated maps have been proposed as a potential intuitive and appealing technique for the visualization of information in a dynamic way, particularly for the detection of spatio-temporal data relations. We aim to address these issues by presenting a comparative study between static and animated representations of human movement on a mobile device context. Our results suggest that although it may not significantly improve user understanding of the data, the use of animated maps is a preferred and less interactively demanding option over static maps.
Tiled Depth of Field Splatting
(The Eurographics Association, 2016) Selgrad, Kai; Franke, Linus; Stamminger, Marc; Luis Gonzaga Magalhaes and Rafal Mantiuk
We present a method to compute post-processing depth of field (DOF) that produces more accurate results than previous approaches. Our method is based on existing approaches, namely DOF rendering by splatting and fast, tile-based particle accumulation. Using tile-based accumulation allows us to correctly sort out of focus pixels and apply proper alpha-blending to avoid artifacts commonly encountered with filter-based depth of field methods.
Buoyancy Optimization for Computational Fabrication
(The Eurographics Association and John Wiley & Sons Ltd., 2016) Wang, Lingfeng; Whiting, Emily; Joaquim Jorge and Ming Lin
This paper introduces a design and fabrication pipeline for creating floating forms. Our method optimizes for buoyant equilibrium and stability of complex 3D shapes, applying a voxel-carving technique to control the mass distribution. The resulting objects achieve a desired floating pose defined by a user-specified waterline height and orientation. In order to enlarge the feasible design space, we explore novel ways to load the interior of a design using prefabricated components and casting techniques. 3D printing is employed for high-precision fabrication. For larger scale designs we introduce a method for stacking lasercut planar pieces to create 3D objects in a quick and economic manner. We demonstrate fabricated designs of complex shape in a variety of floating poses.
Multi-Focus Plenoptic Simulator and Lens Pattern Mixing for Dense Depth Map Estimation
(The Eurographics Association, 2016) Ferreira, Rodrigo; Cunha, Joel; Goncalves, Nuno; T. Bashford-Rogers and L. P. Santos
Light field cameras capture a scene's multi-directional light field with one image, allowing the estimation of depth. In this paper, we introduce a fully automatic method for depth estimation from a single plenoptic image running a RANSAC-like algorithm for feature matching. The novelty about our method is the use of different focal-length lenses for multiple depth map refining, generating a dense depth map for future all-in-focus renders. We also present a plenoptic simulator which produces a plenoptic dataset from a 3D computer rendered scene. This simulator, which is unique, as far as we known, allows testing of plenoptic oriented algorithms since it can reproduce datasets with desired scene characteristics, providing the depth ground truth for error measurement. This work is a on-going project with promising results.
Texel Shading
(The Eurographics Association, 2016) Hillesland, Karl E.; Yang, J. C.; T. Bashford-Rogers and L. P. Santos
We have developed a texture space shading system built on modern graphics hardware. It begins with a conventional rasterization stage, but records texel accesses as shading work rather than running a shade per pixel. Shading is performed by a separate compute stage, storing the results in a texture. As a baseline, the texels correspond to those required for mipmapped texturing. A final stage collects data from the texture. Storing results in a texture allows for reuse across frames. We also show how adapting shade rate to less than once per pixel further increases performance. We vary shading load to show when these techniques provide a performance win, with up to 4.1x speedup in our experiments at shading times less than 4 ms.

Browse

Browsing EG2016 by Issue Date

Results Per Page

Sort Options