Top-DB-Net: Top DropBlock for Activation Enhancement in Person Re-Identification

12 Oct 2020  ยท  Rodolfo Quispe, Helio Pedrini ยท

Person Re-Identification is a challenging task that aims to retrieve all instances of a query image across a system of non-overlapping cameras. Due to the various extreme changes of view, it is common that local regions that could be used to match people are suppressed, which leads to a scenario where approaches have to evaluate the similarity of images based on less informative regions. In this work, we introduce the Top-DB-Net, a method based on Top DropBlock that pushes the network to learn to focus on the scene foreground, with special emphasis on the most task-relevant regions and, at the same time, encodes low informative regions to provide high discriminability. The Top-DB-Net is composed of three streams: (i) a global stream encodes rich image information from a backbone, (ii) the Top DropBlock stream encourages the backbone to encode low informative regions with high discriminative features, and (iii) a regularization stream helps to deal with the noise created by the dropping process of the second stream, when testing the first two streams are used. Vast experiments on three challenging datasets show the capabilities of our approach against state-of-the-art methods. Qualitative results demonstrate that our method exhibits better activation maps focusing on reliable parts of the input images.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Person Re-Identification CUHK03 detected Top-DB-Net MAP 74.2 # 8
Rank-1 77.3 # 8
Person Re-Identification CUHK03 detected Top-DB-Net + RK MAP 86.9 # 1
Rank-1 85.7 # 1
Person Re-Identification CUHK03 labeled Top-DB-Net + RK MAP 88.5 # 2
Rank-1 86.7 # 4
Person Re-Identification CUHK03 labeled Top-DB-Net MAP 75.4 # 11
Rank-1 79.4 # 9
Person Re-Identification DukeMTMC-reID Top-DB-Net Rank-1 87.5 # 48
mAP 73.5 # 57
Person Re-Identification DukeMTMC-reID Top-DB-Net + RK Rank-1 90.9 # 23
mAP 88.6 # 16
Person Re-Identification Market-1501 Top-DB-Net Rank-1 94.9 # 61
mAP 85.8 # 72
Person Re-Identification Market-1501 Top-DB-Net + RK Rank-1 95.5 # 42
mAP 94.1 # 17
Person Re-Identification Market-1501-C TDB Rank-1 28.56 # 19
mAP 8.90 # 19
mINP 0.20 # 20

Methods