|
|
Object Detection Based on Human Visual Perception |
|
|
Object detection is essential for intelligent visual surveillance, remote sensing, traffic monitoring, and even for context-aware applications. State-of-the-art approaches can achieve higher detection quality by utilizing context dependant information either explicitly or implicitly. Consequently, these approaches are highly instable for dynamic unconstraint environments, as well as for abnormality detection in the context itself. This talk will introduce a novel object detection technique for achieving high stability across unconstrained environments by incorporating human perceptual characteristics in the underlying statistical model. |
|
|
Mahfuzul Haque is a PhD Student of GSIT, Monash University. Currently, he is working in the area of intelligent visual surveillance focusing on background modelling, object detection, scene event detection, and scene behaviour analysis. Prior to Monash, he played different roles as a professional Software Engineer in Bangladesh and Switzerland. He holds a BS degree in Computer Science from BUET and a member of the IEEE. His personal web address is: http://www.mahfuzulhaque.com. |
|
