Video Anomaly Detection With Probabilistic Modelling and Ensemble Learning on Deep Spatiotemporal Features
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
ABSTRACT Video surveillance systems are commonly employed to monitor activities and ensure the safety and security of various environments. Anomaly detection in these systems is challenging due to the rarity and high variability of abnormal events. Integrating anomaly detection enables the identification of atypical or suspicious activities. This paper proposes a novel approach for video anomaly detection based on ensemble learning in a weakly supervised setting. The method consists of a two‐stage framework. In the first stage, spatiotemporal features are extracted from video data using 3D deep networks, followed by a multi‐scale attention module to enhance feature representation. Anomalous events are then identified by analysing discrepancies in probabilistic distributions, incorporating multi‐instance learning with a novel term in the loss function. In the second stage, the detection process is refined through ensemble learning strategies to optimise overall performance. The effectiveness of the proposed framework is demonstrated through extensive experiments on five benchmark datasets: UCF‐Crime, XD‐Violence, ShanghaiTech, CUHK Avenue, and UCSD Ped2. The method achieves frame‐level AUC scores of 97.89% on ShanghaiTech, 95.97% on CUHK Avenue, 97.38% on UCSD Ped2, 94.02 on XD‐Violence, and 80.86% on UCF‐Crime, showing competitive performance and highlighting the potential of ensemble‐based weakly supervised methods for video anomaly detection.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it