Research Project Title:
Parallel High-Efficiency Video Coding
abstract:With the growing popularity of high-pixel density displays comes increased demand for high-resolution video content. Such content can't be streamed or stored practically without video-coding algorithms that offer both high performance and compression. HEVC, also known as H.265, is a video- compression standard that offers one of the best compression ratios. However, the current implementation doesn't take full advantage of parallelism, and efforts to do so have been hampered by race conditions and other bugs. This project aims to resolve these bugs to create the fastest, most correct implementation of HEVC and to use the process of debugging HEVC as a source of inspiration for tools that simplify parallelizing code in general.
"I'm doing a SuperUROP project because I want more research experience. I've done some parallel programming in my personal projects, but I hope this research will teach me much more about parallelization and the current state-of-the-art tools used. I'm excited about this project because it has the potential to help both the experience of the average video watcher and the workflow of developers."