DIRT: deferred image-based ray tracing

We introduce a novel approach to image-space ray tracing ideally suited for the photorealistic synthesis of fully dynamic environments at interactive frame rates. Our method, designed entirely on the rasterization pipeline, alters the acceleration data structure construction from a per-fragment to a per-primitive basis in order to simultaneously support three important, generally conflicting in prior art, objectives: fast construction times, analytic intersection tests and reduced memory requirements. In every frame, our algorithm operates in two stages: A compact representation of the scene geometry is built based on primitive linked-lists, followed by a traversal step that decouples the ray-primitive intersection tests from the illumination calculations; a process inspired by deferred rendering and the path integral formulation of light transport. Efficient empty space skipping is achieved by exploiting several culling optimizations both in xy- and z-space, such as pixel frustum clipping, depth subdivision and lossless buffer down-scaling. An extensive experimental study is finally offered showing that our method advances the area of image-based ray tracing under the constraints posed by arbitrarily complex and animated scenarios.

[1]  Carsten Dachsbacher,et al.  Object-Order Ray Tracing for Fully Dynamic Scenes , 2014 .

[2]  P. J. Narayanan,et al.  Raytracing Dynamic Scenes on the GPU Using Grids , 2012, IEEE Transactions on Visualization and Computer Graphics.

[3]  Vlastimil Havran,et al.  Performance Comparison of Bounding Volume Hierarchies and Kd‐Trees for GPU Ray Tracing , 2016, Comput. Graph. Forum.

[4]  Daniel Kopta,et al.  Fast, effective BVH updates for animated scenes , 2012, I3D '12.

[5]  Kirill Garanzha,et al.  Simpler and faster HLBVH with work queues , 2011, HPG '11.

[6]  Jacopo Pantaleoni,et al.  HLBVH: hierarchical LBVH construction for real-time ray tracing of dynamic geometry , 2010, HPG '10.

[7]  Tero Karras,et al.  Maximizing parallelism in the construction of BVHs, octrees, and k-d trees , 2012, EGGH-HPG'12.

[8]  Warren A. Hunt,et al.  The Visibility Buffer: A Cache-Friendly Approach to Deferred Shading , 2013 .

[9]  Georgios Papaioannou,et al.  A multiview and multilayer approach for interactive ray tracing , 2016, I3D.

[10]  Derek Nowrouzezahrai,et al.  Fast Global Illumination Approximations on Deep G-Buffers , 2014 .

[11]  Kirill Garanzha The Use of Precomputed Triangle Clusters for Accelerated Ray Tracing in Dynamic Scenes , 2009, Comput. Graph. Forum.

[12]  Ingo Wald,et al.  State of the Art in Ray Tracing Animated Scenes , 2009, Comput. Graph. Forum.

[13]  Timo Aila,et al.  Fast parallel construction of high-quality bounding volume hierarchies , 2013, HPG '13.

[14]  Andreas Dietrich,et al.  OptiX: a general purpose ray tracing engine , 2010, SIGGRAPH 2010.

[15]  Vlastimil Havran,et al.  Heuristic ray shooting algorithms , 2000 .

[16]  David K. McAllister,et al.  OptiX: a general purpose ray tracing engine , 2010, ACM Trans. Graph..

[17]  Fan Zhang,et al.  Ray tracing via GPU rasterization , 2014, The Visual Computer.

[18]  Michael Goesele,et al.  An adaptive acceleration structure for screen-space ray tracing , 2015, HPG '15.

[19]  Philipp Slusallek,et al.  Two‐Level Grids for Ray Tracing on GPUs , 2011, Comput. Graph. Forum.

[20]  Ioannis Fudos,et al.  Depth-Fighting Aware Methods for Multifragment Rendering , 2013, IEEE Transactions on Visualization and Computer Graphics.

[21]  J. Bittner,et al.  T‐SAH: Animation Optimized Bounding Volume Hierarchies , 2015, Comput. Graph. Forum.

[22]  Michael Wimmer,et al.  CHC+RT: Coherent Hierarchical Culling for Ray Tracing , 2015, Comput. Graph. Forum.

[23]  Carsten Dachsbacher,et al.  Deferred attribute interpolation for memory-efficient deferred shading , 2015, HPG '15.

[24]  Michael Mara,et al.  Efficient GPU Screen-Space Ray Tracing , 2014 .

[25]  Michael C. Doggett,et al.  Real-time multiply recursive reflections and refractions using hybrid rendering , 2014, The Visual Computer.