Fast analytical motion blur with transparency

We introduce a practical parallel technique to achieve real-time motion blur for textured and semitransparent triangles with high accuracy using modern commodity GPUs. In our approach, moving triangles are represented as prisms. Each prism is bounded by the initial and final position of the triangle during one animation frame and three bilinear patches on the sides. Each prism covers a number of pixels for a certain amount of time according to its trajectory on the screen. We efficiently find, store and sort the list of prisms covering each pixel including the amount of time the pixel is covered by each prism. This information, together with the color, texture, normal, and transparency of the pixel, is used to resolve its final color. We demonstrate the performance, scalability, and generality of our approach in a number of test scenarios, showing that it achieves a visual quality practically indistinguishable from the ground truth in a matter of just a few milliseconds, including rendering of textured and transparent objects. A supplementary video has been made available online. 1 © 2021 Elsevier Ltd. All rights reserved.

[1]  Eric Enderton,et al.  Real-time stochastic rasterization on conventional GPU architectures , 2010, HPG '10.

[2]  T. Akenine-Möller,et al.  Analytical motion blur rasterization with compression , 2010, High Performance Graphics.

[3]  I. Wald,et al.  Watertight Ray/Triangle Intersection , 2013 .

[4]  Hans-Peter Seidel,et al.  Deep Shading: Convolutional Neural Networks for Screen Space Shading , 2016, Comput. Graph. Forum.

[5]  Derek Nowrouzezahrai,et al.  High Performance Non-linear Motion Blur , 2015, EGSR.

[6]  Elmar Eisemann,et al.  Adding Motion Blur to Still Images , 2018, Graphics Interface.

[7]  Pat Hanrahan,et al.  Space-time hierarchical occlusion culling for micropolygon rendering with motion blur , 2010, HPG '10.

[8]  Pat Hanrahan,et al.  Hardware implementation of micropolygon rasterization with motion and defocus blur , 2010, HPG '10.

[9]  Pat Hanrahan,et al.  Data-parallel rasterization of micropolygons with defocus and motion blur , 2009, High Performance Graphics.

[10]  Daniel Kopta,et al.  Time Interval Ray Tracing for Motion Blur , 2018, IEEE Transactions on Visualization and Computer Graphics.

[11]  Morgan McGuire,et al.  A reconstruction filter for plausible motion blur , 2012, I3D '12.

[12]  Tom Duff,et al.  Compositing digital images , 1984, SIGGRAPH.

[13]  Alexander Reshetov.  Cool Patches: A Geometric Approach to Ray/Bilinear Patch Intersections , 2019, Ray Tracing Gems.

[14]  Kyoungsu Oh,et al.  Real-time motion blur using extruded triangles , 2017, Multimedia Tools and Applications.

[15]  Robert Bridson,et al.  Efficient geometrically exact continuous collision detection , 2012, ACM Trans. Graph..

[16]  Weifeng Liu,et al.  Fast segmented sort on GPUs , 2017, ICS.

[17]  Kurt Akeley,et al.  The accumulation buffer: hardware support for high-quality rendering , 1990, SIGGRAPH.

[18]  João Luiz Dihl Comba,et al.  Memory-Efficient Order-Independent Transparency with Dynamic Fragment Buffer , 2012, 2012 25th SIBGRAPI Conference on Graphics, Patterns and Images.

[19]  Elmar Eisemann,et al.  Controllable Motion-Blur Effects in Still Images , 2020, IEEE Transactions on Visualization and Computer Graphics.

[20]  Tomas Akenine-Möller,et al.  Backface Culling for Motion Blur and Depth of Field , 2010, J. Graphics, GPU, & Game Tools.

[21]  Derek Nowrouzezahrai,et al.  A Fast and Stable Feature-Aware Motion Blur Filter , 2014, High Performance Graphics.

[22]  Tomas Akenine-Möller,et al.  Stochastic rasterization using time-continuous triangles , 2007, GH '07.

[23]  Norman I. Badler,et al.  Temporal anti-aliasing in computer generated animation , 1983, SIGGRAPH.

[24]  Aaron E. Lefohn,et al.  Adaptive image space shading for motion and defocus blur , 2012, EGGH-HPG'12.

[25]  Tomas Akenine-Möller,et al.  Theory and analysis of higher-order motion blur rasterization , 2013, HPG '13.

[26]  Markus H. Gross,et al.  Controlling Motion Blur in Synthetic Long Time Exposures , 2019, Comput. Graph. Forum.

[27]  Jonathan T. Barron,et al.  Learning to Synthesize Motion Blur , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).