Depth gradient based region of interest generation for pedestrian detection
Maral Mesmakhosroshahi, Kwanghoon Chung, Yun‐Sik Lee, Joohee Kim · 2014
In this paper, we present a novel region of interest (ROI) generation method for stereo-based pedestrian detection systems. In the proposed algorithm, the vertical gradient of the clustered depth map is used to find the flat regions and variable-sized bounding boxes are used to extract the ROIs on the boundary of these regions. The ROIs are then classified into the pedestrian and non-pedestrian classes. Simulation results show that our proposed algorithm outperforms the existing monocular and stereo-based methods.