Double-pedestrian deformable part model and its application in pedestrian detection
1. Automotive Engineering Research Institute, Jiangsu University, Zhenjiang, Jiangsu 212013, China; 2. School of Automotive and Traffic Engineering, Jiangsu University, Zhenjiang, Jiangsu 212013, China
Abstract：To solve the problem that the traditional pedestrian detection algorithm was prone to missing detection for close or close to each other multi-pedestrian targets, a pedestrian detection algorithm was proposed based on two-pedestrian deformable part model. Model training, model matching and fusion process were included in the algorithm to obtain a double pedestrian deformable part model, and a new method of sub-region matching and fusion was also proposed. Two groups of experiments were designed to compare the effect of single-person deformable part model and double pedestrian deformable part model. The results show that compared with other algorithms, the proposed algorithm is more easy to detect close to each other multi-pedestrian target to effectively reduce missing rate.
蔡英凤1, 刘泽2, 孙晓强1, 陈龙1, 王海2. 双行人可变形部件模型及其在行人检测中的应用[J]. 江苏大学学报（自然科学版）, 2017, 38(4): 379-382.
CAI Ying-Feng-1, LIU Ze-2, SUN Xiao-Qiang-1, CHEN Long-1, WANG Hai-2. Double-pedestrian deformable part model and its application in pedestrian detection[J]. Journal of Jiangsu University(Natural Science Eidtion)
, 2017, 38(4): 379-382.
2017, Vol. 38 
