VadCLIP

Reported on 6 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

3D Anomaly DetectiononUBnormal
AUC-ROC· 2023-08-22
62.32
best: 76.51 (DDRO (SSALA))
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection arXiv:2308.11681
3D Anomaly DetectiononShanghaiTech Weakly Supervised
AUC-ROC· 2023-08-22
97.49
best: 97.91 (DDRO)
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection arXiv:2308.11681
Video Anomaly DetectiononUBnormal
AUC-ROC· 2023-08-22
62.32
best: 76.51 (DDRO (SSALA))
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection arXiv:2308.11681
Video Anomaly DetectiononShanghaiTech Weakly Supervised
AUC-ROC· 2023-08-22
97.49
best: 97.91 (DDRO)
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection arXiv:2308.11681

Methodology2 results

Anomaly DetectiononUBnormal
AUC-ROC· 2023-08-22
62.32
best: 76.51 (DDRO (SSALA))
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection arXiv:2308.11681
Anomaly DetectiononShanghaiTech Weakly Supervised
AUC-ROC· 2023-08-22
97.49
best: 98.14 (PEL)
VadCLIP: Adapting Vision-Language Models for Weakly Supervised Video Anomaly Detection arXiv:2308.11681