Fast pedestrian detection based on object proposals and HOG

Research on pedestrian detection still presents a lot of space for improvements, both on speed and detection accuracy. State-of-the-art object proposals approach has shown the very effective computational efficiency in object detection. In this paper, we present a framework for pedestrian detection based on the object proposals. Instead of scaling the test image to different sizes, we generate a pyramid of HOG templates by simply scaling the primary HOG descriptor. For the test image, Histogram of Oriented Gradient feature vector is calculated only in the region of object proposals without using sliding window, and the corresponding HOG template is selected according to the size of proposal region. Experiment results show that our proposed methods can achieve the equal detection rate while using only 20% time.