Reproduce “Text-Driven Traffic Anomaly Detection with Temporal High-Frequency Modeling in Driving Videos” 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

Rongqin Liang, Yuanman Li, Jiantao Zhou, Xia Li

TTHF detects traffic anomalies in dashcam video by combining representations of consecutive frames and their difference, then aligning video features with language describing normal and abnormal traffic. On DoTA, it reports frame-level AUC of 84.7%, versus 79.3% for STFE; these prior-method scores come from published papers. On DADA-2000, it reports 71.7% “Both” AUC without fine-tuning there.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.