Competitive Fusion Region Proposal Network Based Online Pedestrian Tracking Algorithm

被引：0
|
作者：
Wang B. [1 ,2 ]
Wang Y. [1 ,2 ]
Chen Z. [1 ,2 ]
Yang B. [1 ,2 ]
Gao W. [1 ,2 ]
Wang M. [1 ,2 ]
机构：
[1] College of Information and Electrical Engineering, China Agricultural University, Beijing
[2] Key Laboratory of Agricultural Information Standardization, Ministry of Agriculture and Rural Affairs, China Agricultural University, Beijing
来源：
Nongye Jixie Xuebao/Transactions of the Chinese Society for Agricultural Machinery | 2019年 / 50卷
关键词：
Competitive fusion model; Online learning algorithm; Pedestrian tracking; Region proposal network;
D O I：
10.6041/j.issn.1000-1298.2019.S0.051
中图分类号：
学科分类号：
摘要：
Target tracking is an important part of computer vision, specially the pedestrian detection and tracking is a crucial and difficult field. Many researchers have been devoted to the improvement of target detection and tracking methods. With the wide application of deep convolution network, the result of pedestrian detection and tracking has been improved. However, some complex scenarios are difficult to identify and track by present methods. Therefore, it's necessary to propose an optimal algorithm to improve the performance of pedestrian detection and tracking. The region proposal network, which included multi-layer competitive fusion model was used as pre-training network, and long-term and short-term update strategy in pedestrian tracking task. The pre-training network applied VGG16 to extract feature maps, and then they were put into the multi-layer competitive fusion region proposal network to generate more accurate candidate targets. The online pedestrian tracking algorithm was initialized by the pre-training region proposal network, and the region proposal network was fine-tuned through 500 positive samples and 5 000 false positive examples from the first frame, and then created frame index datasets for long-term and short-term update. Finally, the pedestrian tracking algorithm with continuous updating of region proposal network was accomplished. The model was verified by experiment and in the public datasets named by Caltech, ETH, PETS 2009 and Venice. The test result showed that the region proposal network which included multi-layer competitive fusion model had the perfect performance in pedestrian detection and tracking task, and showed good effects in complex background environment. © 2019, Chinese Society of Agricultural Machinery. All right reserved.
引用
收藏
页码：331 / 338and379
相关论文
共 27 条
  • [11] Kristan M., Matas J., Leonardis A., The visual object tracking vot2015 challenge results, Proceedings of the IEEE International Conference on Computer Vision Workshops, pp. 1-23, (2015)
  • [12] Qu J., Liu Z., Non-background HOG for pedestrian video detection, 2012 8th International Conference on Natural Computation, pp. 535-539, (2012)
  • [13] Yao X., Li X., Zhou J., Pedestrian detection method based on edge symmetry and HOG, Computer Engineering, 38, 5, pp. 179-182, (2012)
  • [14] Chong Y., Kuang H., Li Q., Two-stage pedestrian detection based on multiple features and machine learning, Acta Automatica Sinica, 38, 3, pp. 375-381, (2012)
  • [15] Smeulders A.W.M., Chu D.M., Cucchiara R., Et al., Visual tracking: an experimental survey, IEEE Transactions on Pattern Analysis and Machine Intelligence, 36, 7, pp. 1442-1468, (2013)
  • [16] Ren S., He K., Girshick R., Et al., Faster R-CNN: towards real-time object detection with region proposal networks, Advances in Neural Information Processing Systems, pp. 91-99, (2015)
  • [17] Wen L., Cai Z., Lei Z., Et al., Online spatio-temporal structural context learning for visual tracking, European Conference on Computer Vision, pp. 716-729, (2012)
  • [18] Jia Y., Shelhamer E., Donahue J., Et al., Caffe: convolutional architecture for fast feature embedding, Proceedings of the 22nd ACM International Conference on Multimedia, pp. 675-678, (2014)
  • [19] Wu Y., Lim J., Yang M.H., Online object tracking: a benchmark, Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2411-2418, (2013)
  • [20] Chen Z., Zhang L., Khattak A.M., Et al., Deep feature fusion by competitive attention for pedestrian detection, IEEE Access, 7, pp. 21981-21989, (2019)
123