Dr Rangachary Kommanduri, Visiting Professor of Data Science, SIAS presented a paper titled, ‘Enhanced Semantic-Driven Anomaly Detection: A Multi-Class Strategy for Video Data’ at the 9th International Conference on Computer Vision & Image Processing (CVIP 2024) held at IIITDM Kancheepuram from 19 – 21 December 2024.
The article is co-authored with Dr Mrinmoy Ghorai (Assistant Professor, Indian Institute of Information Technology, Sri City) and proposes the Squeeze-Attentive Encoder Network (SAE-Net), a novel architecture integrating semantic-driven analysis and channel wise feature recalibrated attention for precise multi-class anomaly detection in videos.
