A Robust Multi Descriptor Fusion with One-Class CNN for Detecting Anomalies in Video Surveillance
Bibliographic record
Abstract
In the domain of computer vision and machine learning, Video Anomaly Detection (VAD) has emerged as a pivotal area of inquiry, particularly relevant to security, surveillance, and video analytics.The extraction of pertinent features from video data constitutes a foundational aspect of VAD, enabling the discernment of anomalous patterns and structures.Features such as motion, texture, shape, and aesthetics are extracted and tailored according to the exigencies of the application and the intrinsic properties of the video content.The amalgamation of multiple features is often requisite for refining the accuracy of anomaly detection systems.Given the inherent high dimensionality of video data, dimensionality reduction techniques have been employed to mitigate computational demands and enhance the precision of the anomaly detection process.The present study delineates a novel approach centered on the deployment of a One-Class Convolutional Neural Network (CNN).This network is exclusively trained on normal events to establish a baseline representation of typicality.During the evaluation phase, the network is tasked with predicting the normality or abnormality of new video segments against this established norm.Moreover, this work introduces a novel fused feature descriptor, referred to as the Multiple Feature Descriptor (MFD), which is designed to encapsulate the spatiotemporal attributes of video data effectively.The proposed methodology has been subjected to rigorous testing against publicly available datasets, where it has demonstrated superior performance, outstripping numerous contemporary state-of-the-art methods in both accuracy and computational efficiency.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".