ABSTRACT
Deep images store multiple per-pixel fragments, which include colour and depth information, compared to traditional 2D flat images which store only a single color value. Recently deep images are finding use in an increasing number of applications, including transparency and compositing. Compositing deep images requires merging per-pixel fragment lists in depth order, and little work has been presented on performant approaches.
This paper explores GPU based merging of deep images using different memory layouts for fragment lists --- linked lists, linearised arrays, and a new interleaved arrays approach. We also report performance improvements for merging techniques which leverage GPU memory hierarchy and layout by processing blocks of fragment data using fast registers based on similar techniques used to improve performance of transparency rendering. Our results show a 2--5X improvement from combining these techniques.
- Tom Duff. 2017. Deep Compositing Using Lie Algebras. ACM Transactions on Graphics (TOG) 36, 3 (2017), 26. Google ScholarDigital Library
- Jonathan Egstad, Mark Davis, and Dylan Lacewell. 2015. Improved deep image compositing using subpixel masks. In Proceedings of the 2015 Symposium on Digital Production. ACM, 21--27. Google ScholarDigital Library
- Daniel Heckenberg, Johannes Saam, Colin Doncaster, and Chris Cooper. 2010. Deep compositing. Siggraph. Retrieved from http://www.johannessaam. com/deepImage. pdf (2010).Google Scholar
- Peter Hillman. 2013. The theory of openexr deep samples. Technical Report. Tech. rep., Weta Digital Ltd.Google Scholar
- Pyarelal Knowles, Geoff Leach, and Fabio Zambetta. 2012. Efficient layered fragment buffer techniques. (2012).Google Scholar
- Pyarelal Knowles, Geoff Leach, and Fabio Zambetta. 2014. Fast sorting for exact OIT of complex scenes. The Visual Computer 30, 6-8 (2014), 603--613. Google ScholarDigital Library
- Marilena Maule, Joao LD Comba, Rafael Torchelsen, and Rui Bastos. 2012. Memory-efficient order-independent transparency with dynamic fragment buffer. In 2012 25th SIBGRAPI Conference on Graphics, Patterns and Images. IEEE, 134--141. Google ScholarDigital Library
- Morgan McGuire. 2017. Computer Graphics Archive. (July 2017). https://casual-effects.com/data https://casual-effects.com/data.Google Scholar
- Thomas Porter and Tom Duff. 1984. Compositing digital images. In ACM Siggraph Computer Graphics, Vol. 18. ACM, 253--259. Google ScholarDigital Library
Index Terms
GPU based techniques for deep image merging
Recommendations
Fast raycasting using a compound deep image for VPL range determination
SA '18: SIGGRAPH Asia 2018 Technical BriefsThe concept of using multiple deep images has been explored, under a variety of different names, as a possible acceleration approach for finding ray-geometry intersections. We leverage recent advances in deep image processing from Order Independent ...
Comparing GPU-based multi-volume ray casting techniques
The most essential technique to visualize 3D scalar data is direct volume rendering. For many applications it is necessary that two or more 3D data are visualized simultaneously. We present an overview of data intermixing techniques for visualization ...
Image inpainting based on deep learning: A review
AbstractImage inpainting is an important research direction in the study of computer vision, and is widely used in image editing and photo inpainting etc. Traditional image inpainting algorithms are often difficult to deal with large-scale ...
Highlights- Reviewing the image inpainting on deep learning of the past 15 years.
- ...
Comments