Multi-level Factorisation Net for Person Re-identification

Abstract

Key to effective person re-identification (Re-ID) is modelling discriminative and view-invariant factors of person appearance at both high and low semantic levels. Recently developed deep Re-ID models either learn a holistic single semantic level feature representation and/or require laborious human annotation of these factors as attributes. We propose Multi-Level Factorisation Net (MLFN), a novel network architecture that factorises the visual appearance of a person into latent discriminative factors at multiple semantic levels without manual annotation. MLFN is composed of multiple stacked blocks. Each block contains multiple factor modules to model latent factors at a specific level, and factor selection modules that dynamically select the factor modules to interpret the content of each input image. The outputs of the factor selection modules also provide a compact latent factor descriptor that is complementary to the conventional deeply learned features. MLFN achieves state-of-the-art results on three Re-ID datasets, as well as compelling results on the general object categorisation CIFAR-100 dataset.

Beyond Human Parts: Dual Part-Aligned Representations for Person Re-Identification

“…We apply binary human masks (1 for non-human pixels and 0 for human pixels) to remove the influence of pixels predicted as human parts, which is called as Latent w/o HP. (2) only use human part information within latent part branch. We also apply binary human masks (1 for human pixels and 0 for non-human pixels) to remove the influence of pixels predicted as non-human parts, which is called as Latent w/o NHP.…”

“…The details are as folows: (1) We prepare each mini-batch by randomly sampling 16 classes (identities) and 4 images for each class. (2) We set the weight rate as 1:1 on all three datasets. (3) Given a minibatch of 64 samples, we construct a triplet for each sample by choosing the hardest positive sample and the hardest negative sample measured by their Euclidean distances.…”

VMRFANet: View-specific Multi-Receptive Field Attention Network for Person Re-identification

“…DukeMTMC-reID rank 1 mAP rank 1 mAP SVDNet 82.3 62.1 76.7 56.8 PAN 82.8 63.4 71.6 51.5 MultiScale (Chen et al, 2017) 88.9 73.1 79.2 60.6 MLFN (Chang et al, 2018) 90.0 74.3 81.0 62.8 HA-CNN (Li et al, 2018) 91.2 75.7 80.5 63.8 Mancs (Wang et al, 2018a) 93.1 82.3 84.9 71.8 Attention-Driven (Yang et al, 2019) 94.9 86.4 86.0 74.5 PCB+RPP (Sun et al, 2018) 93.8 81.6 83.3 69.2 HPM (Fu et al, 2018) 94.2 82.7 86.6 74.3 MGN (Wang et al, 2018b) 95.7 86.9 88.7 78.4 VMRFANet(Ours) 95.5 88.1 88.9 80.0 Table 3: Comparison of results on CUHK03-labeled (CUHK03-L) and CUHK03-detected (CUHK03-D) with new protocol (Zhong et al, 2017a). The best results are in bold, while the numbers with underlines denote the second best.…”

“…Model CUHK03-L CUHK03-D rank 1 mAP rank 1 mAP SVDNet 40.9 37.8 41.5 37.3 MLFN (Chang et al, 2018) 54.7 49.2 52.8 47.8 HA-CNN (Li et al, 2018) 44.4 41.0 41.7 38.6 PCB+RPP (Sun et al, 2018) --63.7 57.5 MGN (Wang et al, 2018b) 68.0 67.4 68.0 66.0 MRFANet (Ours) 81.1 78.8 78.9 75.3…”

Intra-Camera Supervised Person Re-Identification: A New Benchmark

“…Supervised person re-id Most existing person re-id models are created by supervised learning methods on a separate set of cross-camera identity labelled training data (Wang et al, 2014b(Wang et al, , 2016bZhao et al, 2017;Chen et al, 2017;Li et al, 2017;Chen et al, 2018b;Li et al, 2018b;Song et al, 2018;Chang et al, 2018;Sun et al, 2018;Shen et al, 2018a;Wei et al, 2018;Hou et al, 2019;Zheng et al, 2019;Zhang et al, 2019;Quan et al, 2019;Zhou et al, 2019). Relying on the strong supervision of cross-camera identity labelled training data, they have achieved remarkable performance boost.…”