Reproduce “Object Segmentation-Assisted Inter Prediction for Versatile Video Coding” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhuoyuan Li, Zikun Yuan, Li Li, Dong Liu, Xiaohu Tang, Feng Wu
SAIP (segmentation-assisted inter prediction) uses masks propagated from reference frames to infer two-region motion partitions, and applies them in motion compensation, motion-vector coding, and encoder-side rate-distortion optimization without separately signaling partition shape. Against VTM-12.0, it reduces luma BD-rate by 0.82%, 0.49%, and 0.37% in Low-delay P, Low-delay B, and Random Access, respectively; gains depend on mask quality and reference motion.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.