</> Code with Papers by Syntology

Papers › Bounding Boxes and Probabilistic Graphical Models: Video Anomaly Detection Simplified

Bounding Boxes and Probabilistic Graphical Models: Video Anomaly Detection Simplified

8 Jul 2024arXiv:2407.06000archive 2025-07-28

Mia Siemon, Thomas B. Moeslund, Barry Norton, Kamal Nasrollahi

In this study, we formulate the task of Video Anomaly Detection as a probabilistic analysis of object bounding boxes. We hypothesize that the representation of objects via their bounding boxes only, can be sufficient to successfully identify anomalous events in a scene. The implied value of this approach is increased object anonymization, faster model training and fewer computational resources. This can particularly benefit applications within video surveillance running on edge devices such as cameras. We design our model based on human reasoning which lends itself to explaining model output in human-understandable terms. Meanwhile, the slowest model trains within less than 7 seconds on a 11th Generation Intel Core i9 Processor. While our approach constitutes a drastic reduction of problem feature space in comparison with prior art, we show that this does not result in a reduction in performance: the results we report are highly competitive on the benchmark datasets CUHK Avenue and ShanghaiTech, and significantly exceed on the latest State-of-the-Art results on StreetScene, which has so far proven to be the most challenging VAD dataset.

PaperPDFCode

Code

milestonesys-research/vad-with-pgms officialmentioned in papermentioned on GitHub report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Anomaly DetectionVideo Anomaly Detection

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Anomaly Detection CUHK Avenue PGM AUC 92.72% #10 of 35 Archive leaderboard report
Anomaly Detection CUHK Avenue PGM RBDC 60.18 #10 of 35 Archive leaderboard report
Anomaly Detection CUHK Avenue PGM TBDC 72.09 #10 of 35 Archive leaderboard report
Anomaly Detection ShanghaiTech PGM AUC 61.28% #30 of 31 Archive leaderboard report
Anomaly Detection ShanghaiTech PGM RBDC 45.40 #30 of 31 Archive leaderboard report
Anomaly Detection ShanghaiTech PGM TBDC 81.87 #30 of 31 Archive leaderboard report
Anomaly Detection Street Scene PGM AUC 72.7 #1 of 1 Archive leaderboard report
Anomaly Detection Street Scene PGM RBDC 30.65 #1 of 1 Archive leaderboard report
Anomaly Detection Street Scene PGM TBDC 66.03 #1 of 1 Archive leaderboard report
Video Anomaly Detection CUHK Avenue PGM AUC 92.72% #2 of 7 Archive leaderboard report
Video Anomaly Detection CUHK Avenue PGM RBDC 60.18 #2 of 7 Archive leaderboard report
Video Anomaly Detection CUHK Avenue PGM TBDC 72.09 #2 of 7 Archive leaderboard report
Video Anomaly Detection ShanghaiTech PGM AUC 61.28% #7 of 7 Archive leaderboard report
Video Anomaly Detection ShanghaiTech PGM RBDC 45.4 #7 of 7 Archive leaderboard report
Video Anomaly Detection ShanghaiTech PGM TBDC 81.87 #7 of 7 Archive leaderboard report
Video Anomaly Detection Street Scene PGM AUC 72.7 #1 of 1 Archive leaderboard report
Video Anomaly Detection Street Scene PGM RBDC 30.65 #1 of 1 Archive leaderboard report
Video Anomaly Detection Street Scene PGM TBDC 66.03 #1 of 1 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections