Method to improve efficiency of human detection using scalemap

• Published in: Electronics letters 2014-02, Vol.50 (4), p.265-267
• Main Authors: Bae, Guntae, Kwak, Sooyeong, Byun, Hyeran, Park, Daeyong
• Format: Article
• Language: English
• Subjects: Accuracy; Applied sciences; Artificial intelligence; camera parameters; Cameras; computational costs; Computational efficiency; Computer science; control theory; systems; Devices; Exact sciences and technology; Human; Image and vision processing and display technology; Image detection; multiscale human detection; object detection; optimal detection scale; Optimization; Pattern recognition. Digital image processing. Computational geometry; real‐world surveillance videos; region of interest estimation; ROI estimation; scale of interest estimation; scalemap; SOI estimation; Surveillance; video surveillance
• ISSN: 0013-5194; 1350-911X; 1350-911X
• DOI: 10.1049/el.2013.3588

An efficient method is introduced for detecting humans in surveillance video. The method improves the performance of multiscale human detection through the use of a scalemap, and does not require knowledge of the camera parameters or the use of additional devices. A scalemap is a map that links each position in the observed image to the optimal detection scale. The proposed method efficiently reduces the computational costs by estimating the scale of interest and the region of interest based on the scalemap, while maintaining the accuracy of the detection. It is experimentally shown through an experiment that the proposed method can improve both the accuracy and the efficiency of real-world surveillance videos.