Download：
Figures of the Article
-
![]()
Illustration of our pipeline. We first apply the channel attention module after the backbone network to explicitly separate the camera-related information from the feature maps. Then, we utilize a new mechanism of dynamically updating the memory dictionary according to the distance between the instance feature and the cluster feature.
-
![]()
The CA module.
-
![]()
Visualization of the attention map in the CA module.
-
![]()
T-SNE visualization of 7 classes in the Market1501 test set between our method with d-up removed (left) and our method (right). We utilize a new mechanism of dynamically updating the memory dictionary according to the distance between the instance and cluster feature.
-
![]()
The effect of batch size.
Related articles
-
2023, 53(9): 0901. DOI: 10.52396/JUSTC-2022-0184
-
2021, 51(5): 390-403. DOI: 10.52396/JUST-2021-0078
-
2018, 48(4): 261-274. DOI: 10.3969/j.issn.0253-2778.2018.04.001
-
2017, 47(2): 99-116. DOI: 10.3969/j.issn.0253-2778.2017.02.002
-
2017, 47(1): 40-47. DOI: 10.3969/j.issn.0253-2778.2017.01.006
-
2015, 45(1): 1-8. DOI: 10.3969/j.issn.0253-2778.2015.01.001
-
2014, 44(4): 292-302. DOI: 10.3969/j.issn.0253-2778.2014.04.006
-
2014, 44(3): 194-202. DOI: 10.3969/j.issn.0253-2778.2014.03.005
-
2012, 42(1): 77-84. DOI: 10.3969/j.issn.0253-2778.2012.01.012
-
2009, 39(7): 769-776.
