Allan F. da Silva
Papers
1
Total Citations
20
H-Index
1
About
Allan F. da Silva is a computer vision researcher whose work focuses on robust anomaly detection in dynamic, real-world environments. His major contribution lies in developing a spatio-temporal codebook approach for identifying unusual events from moving cameras—a significant challenge in surveillance, as most prior methods assumed stationary viewpoints. By encoding both spatial and temporal features, his framework enables reliable detection of anomalies such as erratic pedestrian behavior or vehicle motion, even as the camera itself pans or tilts. This work, published in 2017, has garnered 20 citations, reflecting its practical relevance for autonomous monitoring systems. da Silva’s research bridges the gap between theoretical pattern recognition and applied video analytics, offering a scalable solution for smart city and security applications. His approach has been noted for its computational efficiency, making it suitable for real-time deployment on embedded platforms. For students and researchers in computer vision, da Silva’s work exemplifies how careful feature engineering can overcome the limitations of traditional static-camera models, opening new avenues for adaptive surveillance in uncontrolled settings.
Research Focus
Key Achievements
Top Papers
- 1Anomaly detection with a moving camera using spatio-temporal codebooks20 citations · 2017