Fast pedestrian detection with a cascade of multi-Hogs

Accurate and efficient human detection has become an important area for research in computer vision. In order to solve problems in the past human detection algorithms such as features with fixed sizes, fixed positions and fixed number, the human detection based on multi-Hogs algorithm was proposed. A novel representation of irregular rectangles called Vertex-Vector Representation was introduced and generation algorithms of Vertex-Vector Representation and inner points of a new rectangle in the intersection of two irregular rectangles were proposed. Through intersection tests and feature integration, the algorithm can dynamically generate the features closer to human body contours. Basically maintaining the detection speed, the detection accuracy was improved by our algorithm.