Laser & Optoelectronics Progress, Volume. 61, Issue 8, 0837006(2024)
Template Update Mechanism for Single Target Tracking Incorporating Memory Information
Fig. 1. Tracking algorithm based on Siamese architecture
Fig. 2. Tracking algorithm with the template UM
Fig. 3. Structure of MFM
Fig. 4. Probability of response map and the tracking result. (a) Probability of response map; (b) tracking result
Fig. 5. Structure of WFM
Fig. 6. Visualization results of different video sequences for the VOT2019 dataset. (a1)‒(a4) Original images of girl; (b1)‒(b4) tracking results of girl; (c1)‒(c4) local zoomed images of girl's tracking results; (d1)‒(d4) original images of gymnastics1; (e1)‒(e4) tracking results of gymnastics1; (f1)‒(f4) local zoomed images of gymnastics1's tracking results; (g1)‒(g4) original images of tiger; (h1)‒(h4) tracking results of tiger; (i1)‒(i4) local zoom images of tiger's tracking results; (j1)‒(j4) original images of nature; (k1)‒(k4) tracking results of nature; (l1)‒(l4) local zoomed images of nature's tracking results
|
|
|
|
|
Get Citation
Copy Citation Text
Yuwen Mao, Baozhen Ge, Jianing Quan, Qibo Chen. Template Update Mechanism for Single Target Tracking Incorporating Memory Information[J]. Laser & Optoelectronics Progress, 2024, 61(8): 0837006
Category: Digital Image Processing
Received: Jun. 19, 2023
Accepted: Jul. 24, 2023
Published Online: Mar. 20, 2024
The Author Email: Ge Baozhen (gebz@tju.edu.cn)