Reproduce “Selective Event Evidence and Persistent Temporal Priors for Event-Based Point Tracking” 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
Xiaokai Song, Zhang Li, Leqi Liu, Yibin Ye, Kun Wang, Xichao Teng
SEPT extends ETAP, an event-only tracker, by weighting events before feature extraction and carrying a geometric prior across tracking windows. The prior uses a fixed quadratic model, while learned weighting changes evidence used for matching. On EVIMO2 without fine-tuning, SEPT raises average Jaccard from 0.635 to 0.673, while visibility accuracy falls from 0.943 to 0.936.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.