Acta Optica Sinica, Volume. 39, Issue 9, 0915001(2019)
Correlation Filter Tracking Based on Adaptive Feature Fusion and Model Updating
Figures & Tables(7)
Fig. 3. 179th frame and corresponding response maps of singer2 video. (a) 179th frame; (b) MIX; (c) HOG; (d) fusion
Fig. 4. Maximum response value, adaptive weight, and partial frames in David3. (a) Maximum response map and adaptive weight; (b) partial frames
Fig. 5. Comparison of distance precision plots and success rate plots of eight tracking algorithms on 36 color sequences. (a) Distance precision plots; (b) success rate plots
Fig. 6. Tracking results of six tracking algorithms. (a) Couple; (b) Lemming; (c) Liquor; (d) Girl; (e) Motor Rolling
Table 1. Average tracking performance indexes of eight tracking algorithms on 36 color sequences
View tableTable 1. Average tracking performance indexes of eight tracking algorithms on 36 color sequences
Algorithm Proposed CCOT SAMF Staple KCF DSST CN CSK Mean DP /% 85.4 83.2 79.0 78.2 77.0 76.7 67.8 55.2 Mean OP /% 79.2 78.0 72.0 75.2 63.8 68.9 52.9 41.8 Mean FPS /(frame·s-1) 28.39 31.42 17.53 29.26 149.60 28.12 63.23 264.33
Tools
Get Citation
Copy Citation Text
Min Chang, Kai Shen, Xuedian Zhang, Jia Du, Feng Li. Correlation Filter Tracking Based on Adaptive Feature Fusion and Model Updating[J]. Acta Optica Sinica, 2019, 39(9): 0915001
Paper Information
Category: Machine Vision
Received: Mar. 5, 2019
Accepted: May. 5, 2019
Published Online: Sep. 9, 2019
The Author Email: Chang Min (changmin@usst.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20