11 papers
Frame-Level Evaluation in Weakly Supervised Video Anomaly Detection Mostly Measures Video-Level Ranking
Inpyo Song, Jangwon Lee
Weakly supervised video anomaly detectors are trained with video-level labels but are commonly evaluated as temporal localizers using Micro-AUROC or AP over pooled test frames. Bec…
A VLM Answer Is Not an Anomaly Score: Rank Compression in Training-Free Video Anomaly Detection
Inpyo Song, Jangwon Lee
Vision-language models enable training-free video anomaly detection by answering questions about video segments. VAD benchmarks, however, require a scalar anomaly score for each se…
Rethinking Open-World Video Anomaly Detection: Diagnosing Definition Blindness
Inpyo Song, Jangwon Lee
Open-world video anomaly detection (OWVAD) is expected to detect events that match a user-specified definition of abnormality. This requirement is stronger than generic anomaly loc…
Bounding-Box Trajectories Matter for Video Anomaly Detection
Inpyo Song, Jangwon Lee
Video anomaly detection is critical for public safety and security, yet remains highly challenging despite extensive research due to large variations in appearance, viewpoint, and…
Instance-Aligned Captions for Explainable Video Anomaly Detection
Inpyo Song, Minjun Joo, Joonhyung Kwon +2
Explainable video anomaly detection (VAD) is crucial for safety-critical applications, yet even with recent progress, much of the research still lacks spatial grounding, making the…
PCEval: A Benchmark for Evaluating Physical Computing Capabilities of Large Language Models
Inpyo Song, Eunji Jeon, Jangwon Lee
Large Language Models (LLMs) have demonstrated remarkable capabilities across various domains, including software development, education, and technical assistance. Among these, sof…