DSpace Repository

Search Results

Now showing 1 - 10 of 39

Compressed Bounding Volume Hierarchies for Efficient Ray Tracing of Disperse Hair
(The Eurographics Association, 2018) Martinek, Magdalena; Stamminger, Marc; Binder, Nikolaus; Keller, Alexander; Beck, Fabian and Dachsbacher, Carsten and Sadlo, Filip
Ray traced human hair is becoming more and more ubiquitous in photorealistic image synthesis. Despite hierarchical data structures for accelerated ray tracing, performance suffers from the bad separability inherent with ensembles of hair strands. We propose a compressed acceleration data structure that improves separability by adaptively subdividing hair fibers. Compression is achieved by storing quantized as well as oriented bounding boxes and an indexing scheme to specify curve segments instead of storing them. We trade memory for speed, as our approach may use more memory, however, in cases of highly curved hair we can double the number of traversed rays per second over prior work. With equal memory we still achieve a speed-up of up to 30%, with equal performance we can reduce memory by up to 30%.
Projection Mapping for In-Situ Surgery Planning by the Example of DIEP Flap Breast Reconstruction
(The Eurographics Association, 2021) Martschinke, Jana; Klein, Vanessa; Kurth, Philipp; Engel, Klaus; Ludolph, Ingo; Hauck, Theresa; Horch, Raymund; Stamminger, Marc; Oeltze-Jafra, Steffen and Smit, Noeska N. and Sommer, Björn and Nieselt, Kay and Schultz, Thomas
Nowadays, many surgical procedures require preoperative planning, mostly relying on data from 3D imaging techniques like computed tomography or magnetic resonance imaging. However, preoperative assessment of this data is carried out on the PC (using classical CT/MR viewing software) and not on the patient's body itself. Therefore, surgeons need to transfer both their overall understanding of the patient's individual anatomy and also specific markers and labels for important points from the PC to the patient only with the help of imaginative power or approximative measurement. In order to close the gap between preoperative planning on the PC and surgery on the patient, we propose a system to directly project preoperative knowledge to the body surface by projection mapping. As a result, we are able to display both assigned labels and a volumetric and view-dependent view of the 3D data in-situ. Furthermore, we offer a method to interactively navigate through the data and add 3D markers directly in the projected volumetric view. We demonstrate the benefits of our approach using DIEP flap breast reconstruction as an example. By means of a small pilot study, we show that our method outperforms standard surgical planning in accuracy and can easily be understood and utilized even by persons without any medical knowledge.
Path-Traced Motion Blur using Motion Trees
(The Eurographics Association, 2020) Martinek, Magdalena; Thiemann, Philip; Stamminger, Marc; Biasotti, Silvia and Pintus, Ruggero and Berretti, Stefano
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 particular, for fast moving objects, undersampling in the time domain results in severe artefacts. In this paper, we propose the Motion Tree, a novel Level-of-Detail data structure for efficient handling of animated objects, that both filters in the spatial and the temporal domain. The Motion Tree is a compact nesting of a temporal interval binary tree for filtering time consecutive data and a sparse voxel octree (SVO) which simplifies spatially nearby data. It is generated during a pre-process and fits nicely into any conventional physically based path tracer. When used in a production-scale environment it significantly reduces memory requirements allowing for a speedup in rendering performance with user control over the degree of impact on quality.
Topological Triangle Sorting for Predefined Camera Paths
(The Eurographics Association, 2016) Weber, Christoph; Stamminger, Marc; Matthias Hullin and Marc Stamminger and Tino Weinkauf
We present a preprocessing pipeline for triangle meshes that topologically sorts all triangles for a given camera and scene animation in front-to-back or back-to-front order. This allows us to efficiently render a given animation without depth buffer, and to include transparency. We also remove non-contributing triangles, thus improving render time, especially when applying anti-aliasing. To this end we first record the visible triangles of a sequence of frames. For every frame we create a directed graph storing occlusion information. After a topological sort of this graph, all triangles are sorted properly. The contribution of this paper is the reduction of redundancy by merging the graphs of all frames. The result of our pipeline is a single sorted index buffer, over which we slide a window that yields sorted index buffers for each single frame. Circular dependencies are broken by placing duplicates of the affected triangles in the index buffer. Our sliding window then displays only frame specific triangles in their proper order. We conclude by demonstrating the benefits of removing invisible triangles and disabling the hardware visibility test.
On Floating-Point Normal Vectors
(The Eurographics Association and Blackwell Publishing Ltd, 2010) Meyer, Quirin; Suessmuth, Jochen; Sussner, Gerd; Stamminger, Marc; Greiner, Guenther
In this paper we analyze normal vector representations. We derive the error of the most widely used representation, namely 3D floating-point normal vectors. Based on this analysis, we show that, in theory, the discretization error inherent to single precision floating-point normals can be achieved by 250.2 uniformly distributed normals, addressable by 51 bits. We review common sphere parameterizations and show that octahedron normal vectors perform best: they are fast and stable to compute, have a controllable error, and require only 1 bit more than the theoretical optimal discretization with the same error.
Neural Volumetric Level of Detail for Path Tracing
(The Eurographics Association, 2024) Stadter, Linda; Hofmann, Nikolai; Stamminger, Marc; Linsen, Lars; Thies, Justus
We introduce a neural level of detail pipeline for use in a GPU path tracer based on a sparse volumetric representation derived from neural radiance fields. We pre-compute lighting and occlusion to train a neural radiance field which faithfully captures appearance and shading via image-based optimization. By converting the resulting neural network into an efficiently rendered representation, we eliminate costly evaluations at runtime and keep performance competitive. When applying our representation to certain areas of the scene, we trade a slight bias from gradient-based optimization and lossy volumetric conversion for highly anti-aliased results at low sample counts. This enables virtually noise-free and temporally stable results at low computational cost and without any additional post-processing, such as denoising. We demonstrate the applicability of our method to both individual objects and a challenging outdoor scene composed of highly detailed foliage.
Data‐Parallel Decompression of Triangle Mesh Topology
(The Eurographics Association and Blackwell Publishing Ltd., 2012) Meyer, Quirin; Keinert, Benjamin; Sußner, Gerd; Stamminger, Marc; Holly Rushmeier and Oliver Deussen
We propose a lossless, single‐rate triangle mesh topology codec tailored for fast data‐parallel GPU decompression. Our compression scheme coherently orders generalized triangle strips in memory. To unpack generalized triangle strips efficiently, we propose a novel parallel and scalable algorithm. We order vertices coherently to further improve our compression scheme. We use a variable bit‐length code for additional compression benefits, for which we propose a scalable data‐parallel decompression algorithm. For a set of standard benchmark models, we obtain (min: 3.7, med: 4.6, max: 7.6) bits per triangle. Our CUDA decompression requires only about 15% of the time it takes to render the model even with a simple shader.We propose a lossless, single‐rate triangle mesh topology codec tailored for fast data‐parallel GPU decompression. Our compression scheme coherently orders generalized triangle strips in memory. To unpack generalized triangle strips efficiently, we propose a novel parallel and scalable algorithm. We order vertices coherently to further improve our compression scheme. We use a variable bit‐length code for additional compression benefits, for which we propose a scalable data‐parallel decompression algorithm. For a set of standard benchmark models, we obtain (min: 3.7, med: 4.6, max: 7.6) bits per triangle. Our CUDA decompression requires only about 15% of the time it takes to render the model even with a simple shader.
Professional Board Report
(2024-04-18) Stamminger, Marc
Faster Ray-Traced Shadows for Hybrid Rendering of Fully Dynamic Scenes by Pre-BVH Culling
(The Eurographics Association, 2015) Selgrad, Kai; Müller, Jonas; Stamminger, Marc; Andrea Giachetti and Silvia Biasotti and Marco Tarini
With ever increasing ray traversal and hierarchy construction performance the application of ray tracing to problems often tackled by rasterization-based algorithms is becoming a viable alternative. This is especially desirable as the ground truth for these algorithms is often determined by using ray tracing and thus directly applying it is the simplest way to generate images satisfying the reference. In this paper we propose a very efficient pre-process to speed up the construction and traversal of sub-optimal, but fast-to-build hierarchies used for interactive ray tracing and show how it can be applied to shadow rays in a hybrid environment, where ray tracing is used to sample area lights for scene positions found and shaded via rasterization.
Real-time Local Displacement using Dynamic GPU Memory Management
(ACM, 2013) Schäfer, Henry; Keinert, Benjamin; Stamminger, Marc; Kayvon Fatahalian and Christian Theobalt
We propose a novel method for local displacement events in large scenes, such as scratches, footsteps, or sculpting operations. Deformations are stored as displacements for vertices generated by hardware tessellation. Adaptive mesh refinement, application of the displacement and all involved memory management happen completely on the GPU. We show various extensions to our approach, such as on-the-fly normal computation and multi-resolution editing. In typical game scenes we perform local deformations at arbitrary positions in far less than one millisecond. This makes the method particularly suited for games and interactive sculpting applications.

Filters

Author

Subject

Date

Has files

Settings

Sort By

Results per page

Search Results