Research map / Video / Video segmentation and tracking
CLUE-VAD: Structured Semantic Clues for Understanding Explainable Events in Video Anomaly Detection
MYOUNG-CHUL KIM, Junhee Lee, ChaeBeen Bang, MyeongAh Cho
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use all three clues (Action, Environment, Object) instead of two (ablation) | CLUE-VAD · UCF-Crime | AUC (%): 86.89 -> 88.34 | Improved |
| use all three clues instead of a single clue (ablation) | CLUE-VAD · UCF-Crime | AUC (%): 85.82 -> 88.34 | Improved |
| category-aware CLUE weighting instead of self-attention fusion (ablation) | CLUE-VAD Detective module · UCF-Crime | AUC (%): 86.89 -> 88.34 | Improved |
| category-aware CLUE weighting instead of graph-based fusion (ablation) | CLUE-VAD Detective module · UCF-Crime | AUC (%): 86.13 -> 88.34 | Improved |
| category-aware CLUE weighting instead of pairwise inter-CLUE fusion (ablation) | CLUE-VAD Detective module · UCF-Crime | AUC (%): 87.46 -> 88.34 | Improved |
| add visual features to text-only CLUE model (ablation) | CLUE-VAD · UCF-Crime | AUC (%): 88.34 -> 89.47 | Improved |
| add visual features to text-only CLUE model (ablation) | CLUE-VAD · XD-Violence | AP (%): 87.77 -> 88.24 | Improved |
| add textual CLUE features to visual-only model (ablation) | CLUE-VAD · UCF-Crime | AUC (%): 87.46 -> 89.47 | Improved |
| add textual CLUE features to visual-only model (ablation) | CLUE-VAD · XD-Violence | AP (%): 86.35 -> 88.24 | Improved |
| ground InternVL2-2B explanations with CLUE tokens (2B vs 8B InternVL2) | InternVL2 · UCFA anomaly explanation generation | BLEU-4 / CIDEr: BLEU-4 0.0339 -> 0.0739; CIDEr 0.0713 -> 0.0423 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Holmes-VAU — Holmes-VAU: Towards Long-term Video Anomaly Understanding at Any Granularity (CVPR 2025) · better on 2 of 2 shared results
- Ex-VAD — Ex-VAD: Explainable Fine-grained Video Anomaly Detection Based on Visual-Language Models (ICML 2025) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Learning Event Completeness for Weakly Supervised Video Anomaly Detection (ICML 2025 · similarity 0.92)
- Open-Vocabulary Video Anomaly Detection (CVPR 2024 · similarity 0.91)
- Harnessing Large Language Models for Training-free Video Anomaly Detection (CVPR 2024 · similarity 0.91)
- VADTree: Explainable Training-Free Video Anomaly Detection via Hierarchical Granularity-Aware Tree (NeurIPS 2025 · similarity 0.91)
- MoniTor: Exploiting Large Language Models with Instruction for Online Video Anomaly Detection (NeurIPS 2025 · similarity 0.90)
- Ex-VAD: Explainable Fine-grained Video Anomaly Detection Based on Visual-Language Models (ICML 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
