The F-buffer: a rasterization-order FIFO buffer for multi-pass rendering
Proceedings of the ACM SIGGRAPH/EUROGRAPHICS workshop on Graphics hardware
Transparency and Antialiasing Algorithms Implemented with the Virtual Pixel Maps Technique
IEEE Computer Graphics and Applications
Delay streams for graphics hardware
ACM SIGGRAPH 2003 Papers
Proceedings of the 2005 symposium on Interactive 3D graphics and games
GPU-accelerated high-quality hidden surface removal
Proceedings of the ACM SIGGRAPH/EUROGRAPHICS conference on Graphics hardware
Multi-fragment effects on the GPU using the k-buffer
Proceedings of the 2007 symposium on Interactive 3D graphics and games
ACM SIGGRAPH 2007 sketches
Efficient depth peeling via bucket sort
Proceedings of the Conference on High Performance Graphics 2009
Ray casting of multiple volumetric datasets with polyhedral boundaries on manycore GPUs
ACM SIGGRAPH Asia 2009 papers
FreePipe: a programmable parallel rendering architecture for efficient multi-fragment effects
Proceedings of the 2010 ACM SIGGRAPH symposium on Interactive 3D Graphics and Games
Depth-presorted triangle lists
ACM Transactions on Graphics (TOG) - Proceedings of ACM SIGGRAPH Asia 2012
Hi-index | 0.00 |
We present two new image space techniques for efficient rendering of transparent surfaces that exploit partial ordering in the scene geometry. The first technique, called hybrid layer peeling, combines unordered meshes with ordered meshes in an efficient way, and is ideal for scenes such as volumes with embedded transparent meshes. The second technique, called coherent layer peeling, efficiently detects and renders correctly sorted fragment sequences for a given pixel in one iteration, allowing for a smaller number of passes than traditional layer peeling for typical scenes. Although more expensive than hybrid layer peeling by a constant factor, coherent layer peeling applies to a broader class of scenes, including single meshes or collections of meshes. Coherent layer peeling does not require costly clipping or perfect sorting. However, the performance of the algorithm depends on the degree to which the data is sorted. At best, when the data is perfectly sorted, the algorithm renders a correct result in a single iteration. At worst, when the data is sorted in reverse order, the algorithm mimics the performance of layer peeling but with a higher cost per iteration. We conclude with a discussion of a modified form of coherent layer peeling designed for an idealized rasterization architecture that would match layer-peeling in the worst case, while still exploiting correctly sorted sequences when they are present.