• Acta Optica Sinica
  • Vol. 39, Issue 11, 1115002 (2019)
Kuan Yin1, Junli Li1、*, Li Li1, and Chengxi Chu2
Author Affiliations
  • 1College of Computer Science, Sichuan Normal University, Chengdu, Sichuan 610101, China
  • 2Faculty of Electrical Engineering and Computer Science, Ningbo University, Ningbo, Zhejiang 315211, China
  • show less
    DOI: 10.3788/AOS201939.1115002 Cite this Article Set citation alerts
    Kuan Yin, Junli Li, Li Li, Chengxi Chu. Adaptive Feature Update Object-Tracking Algorithm in Complex Situations[J]. Acta Optica Sinica, 2019, 39(11): 1115002 Copy Citation Text show less
    Framework of proposed algorithm
    Fig. 1. Framework of proposed algorithm
    Target-feature visualization. (a) Original image; (b) HOG feature; (c) conv3-4 feature; (d) conv4-4 feature; (e) conv5-4 feature
    Fig. 2. Target-feature visualization. (a) Original image; (b) HOG feature; (c) conv3-4 feature; (d) conv4-4 feature; (e) conv5-4 feature
    Schematic of feature fusion
    Fig. 3. Schematic of feature fusion
    Update mechanism of temporal model
    Fig. 4. Update mechanism of temporal model
    Flow chart of algorithm
    Fig. 5. Flow chart of algorithm
    Qualitative results of 10 tracking algorithms for some video sequences
    Fig. 6. Qualitative results of 10 tracking algorithms for some video sequences
    Tracking results of different feature experts
    Fig. 7. Tracking results of different feature experts
    Tracking accuracy and success rate of algorithm on OTB-2013 and OTB-2015 databases. (a) Tracking accuracy on OTB-2013 database; (b) tracking success rate on OTB-2013 database; (c) tracking accuracy on OTB-2015 database; (d) tracking success rate on OTB-2015 database
    Fig. 8. Tracking accuracy and success rate of algorithm on OTB-2013 and OTB-2015 databases. (a) Tracking accuracy on OTB-2013 database; (b) tracking success rate on OTB-2013 database; (c) tracking accuracy on OTB-2015 database; (d) tracking success rate on OTB-2015 database
    Tracking precision of 11 different attribute video sequences on OTB-2013 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Fig. 9. Tracking precision of 11 different attribute video sequences on OTB-2013 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Tracking success rates of 11 different attribute video sequences on OTB-2013 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Fig. 10. Tracking success rates of 11 different attribute video sequences on OTB-2013 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Tracking precision of 11 different attribute video sequences on OTB-2015 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Fig. 11. Tracking precision of 11 different attribute video sequences on OTB-2015 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Tracking success rates of 11 different attribute video sequences on OTB-2015 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Fig. 12. Tracking success rates of 11 different attribute video sequences on OTB-2015 database. (a) Background clutter; (b) deformation; (c) fast motion; (d) in-plane rotation; (e) illumination variation; (f) low resolution; (g) motion blur; (h) occlusion; (i) out-of-plane rotation; (j) out of view; (k) scale variation
    Tracking precision and success rate of algorithm under long-term tracking. (a) Tracking precision;(b) tracking success rate
    Fig. 13. Tracking precision and success rate of algorithm under long-term tracking. (a) Tracking precision;(b) tracking success rate
    Expert12345678
    Frequency1854188077852534
    Table 1. Frequency statistics of feature experts
    DatabaseParameterOursSTRCFMCCTCF2ECOSRDCFADNetStapleKCFLCT
    OTB-2013Precision0.9020.8890.8830.8910.8550.8380.7980.7820.7400.848
    Success rate0.8760.8450.8630.8090.8060.7890.7210.7380.6230.738
    OTB-2015Precision0.8710.8640.8600.8450.8360.7880.7720.7840.6960.762
    Success rate0.8290.8000.8180.7510.7720.7300.7000.6990.5260.629
    Average FPS3.928.04.21.554.88.012.397.6349.329.4
    Table 2. Tracking accuracy, success rate, and speed of algorithm on OTB-2013 and OTB2015 databases
    Kuan Yin, Junli Li, Li Li, Chengxi Chu. Adaptive Feature Update Object-Tracking Algorithm in Complex Situations[J]. Acta Optica Sinica, 2019, 39(11): 1115002
    Download Citation