aiqbench analysis
This research introduces a Short-Window Sliding Learning framework that leverages LLM-based auto-labeling to create fine-grained datasets from short video clips, significantly enhancing real-time violence detection accuracy. By processing 1-2 second segments instead of long videos, the method preserves temporal continuity for rapid event recognition while achieving high performance on standard benchmarks like RWF-2000 and UCF-Crime. The approach offers a practical pathway for deploying efficient AI models in intelligent surveillance systems without relying on conventional long-video training pipelines.
Short-Window Sliding Learning for Real-Time Violence Detection via LLM-based Auto-Labeling
arXiv:2511.10866v2 Announce Type: replace-cross Abstract: This paper proposes a Short-Window Sliding Learning framework for real-time violence detection in CCTV footages. Unlike conventional long-video training approaches, the proposed method divides videos into 1-2 second clips and applies Large Language Model (LLM)-based auto-caption labeling to construct fine-grained datasets. Each short clip fully utilizes all frames to preserve temporal continuity, enabling precise recognition of rapid violent events. Experiments demonstrate that the proposed method achieves 95.25\% accuracy on RWF-2000 and significantly improves performance on long videos (UCF-Crime: 83.25\%), confirming its strong generalization and real-time applicability in intelligent surveillance systems.
Abstract:This paper proposes a Short-Window Sliding Learning framework for real-time violence detection in CCTV footages. Unlike conventional long-video training approaches, the proposed method divides videos into 1-2 second clips and applies Large Language Model (LLM)-based auto-caption labeling to construct fine-grained datasets. Each short clip fully utilizes all frames to preserve temporal continuity, enabling precise recognition of rapid violent events. Experiments demonstrate that the proposed method achieves 95.25\% accuracy on RWF-2000 and significantly improves performance on long videos (UCF-Crime: 83.25\%), confirming its strong generalization and real-time applicability in intelligent surveillance systems.
5 pages, 2 figures. Accepted paper for the IEIE (Institute of Electronics and Information Engineers) Fall Conference 2025. Presentation on Nov 27, 2025
Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI)
I.2.10; I.4.8; I.2.6
arXiv:2511.10866 [cs.CV]
(or arXiv:2511.10866v2 [cs.CV] for this version)
Full text under the publisher's open licence · 133 words
Related coverage
How Tata Elxsi detects industrial safety risks in seconds on AWS
AWS · Sep 22, 2026
Presentation: Beyond Prompting: Context Engineering and Memory Management for AI Systems at Scale
InfoQ · Jun 10, 2026
How Databricks Feature Store serves features with sub-second freshness
Databricks · Aug 17, 2026
2026.10: You are here 📍
Home Assistant · Oct 7, 2026