@inproceedings{eba9c943fc574515a6181c7eca2dd601,
title = "A new pedestrian detect method in crowded scenes",
abstract = "Most existing pedestrian detection methods always focus on improving detect accuracy of single pedestrian detection, but in this paper we focus on detect crowded pedestrians and recognizing adjacent or overlapped pedestrian exactly. We pro-pose a dissimilarity model to represent difference between adjacent pedestrians by utilizing relative spatial information, body part information, color difference, and crowd density information. Through this model we can accurately distinct every pedestrian in a dense crowd. A deep architecture neural network is used in our model, deep belief network. Its low-level feature learning characteristic makes our model have a more intelligent performance. Some optimization measures are used to make our algorithm more efficient. Experiments on an authority dataset have proved the method's effectiveness.",
keywords = "Crowded scenes, Deep belief network, Dissimilarity, Pedestrian detection",
author = "Hou Xin and Zhang Hong and Yuan Ding",
year = "2013",
doi = "10.1109/GreenCom-iThings-CPSCom.2013.337",
language = "英语",
isbn = "9780769550466",
series = "Proceedings - 2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing, GreenCom-iThings-CPSCom 2013",
pages = "1820--1824",
booktitle = "Proceedings - 2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing, GreenCom-iThings-CPSCom 2013",
note = "2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social Computing, GreenCom-iThings-CPSCom 2013 ; Conference date: 20-08-2013 Through 23-08-2013",
}