Reproduce “TSRN-RTVD: Real-Time Video Deblurring System” 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
Nikita Alutis, Danila Evsyukov, Egor Chistov, Mikhail Voronin, Evgeney Bogatyrev, Dmitriy Vatolin
TSRN-RTVD restores camera-shake-blurred streaming video. It predicts a signal for camera-induced image shift, uses it to align neighboring-frame features, and reconstructs frames from a three-frame window with one-frame look-ahead. On GoPro at 1280 × 720, it reports 30.08 dB PSNR at 30 FPS on RTX 2080 Ti, above the authors’ 25 FPS threshold; this is throughput on the benchmark, not end-to-end latency.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.