End-to-End Comparative Attention Networks for Person Re-Identification

By Hao Liu, Jiashi Feng, Meibin Qi, Jianguo Jiang, Shuicheng Yan. 2016

604 citations recorded by the source corpus.

Person re-identification across disjoint camera views has been widely applied in video surveillance yet it is still a challenging problem. One of the major challenges lies in the lack of spatial and temporal cues, which makes it difficult to deal with large variations of lighting conditions, viewing angles, body poses, and occlusions. Recently, several deep-learning-based person re-identification approaches have been proposed and achieved remarkable performance. However, most of those approaches extract discriminative features from the whole frame at one glimpse without differentiating various parts of the persons to identify. It is essentially important to examine multiple highly discriminative local regions of the person images in details through multiple glimpses for dealing with the large appearance variance. In this paper, we propose a new soft attention-based model, i.e., the end-to-end comparative attention network (CAN), specifically tailored for the task of person re-identification. The end-to-end CAN learns to selectively focus on parts of pairs of person images after taking a few glimpses of them and adaptively comparing their appearance. The CAN model is able to learn which parts of images are relevant for discerning persons and automatically integrates information from different parts to determine whether a pair of images belongs to the same person. In other words, our proposed CAN model simulates the human perception process to verify whether two images are from the same person. Extensive experiments on four benchmark person re-identification data sets, including CUHK01, CHUHK03, Market-1501, and VIPeR, clearly demonstrate that our proposed end-to-end CAN for person re-identification outperforms well established baselines significantly and offer the new state-of-the-art performance.

Subfields: Biomedical Engineering, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Biomedical Engineering

1. Yi Yang
2. Liang Zheng
3. Qi Tian
4. Wanli Ouyang
5. Shiliang Zhang
6. Tao A. Xiang
7. Zhedong Zheng
8. Shaogang Gong

In Computer Vision and Pattern Recognition

1. Yi Yang
2. Luc Van Gool
3. Dacheng Tao
4. Xiaogang Wang
5. Tao A. Xiang
6. Qi Tian
7. Ming-Hsuan Yang
8. Liang Zheng

Citations in the corpus

• DISCO: Dynamic and Invariant Sensitive Channel Obfuscation for deep neural networks (2020, methodology): “While sophisticated learning based partitioning methods have been explored [36, 37, 38, 39], methods such as [40] have achieved outstanding performance with trivial deterministic splitting.”
• Deep Mutual Learning (2017, methodology): “Method ImageNet Single-Query Multi-Query Pretrain? mAP Rank-1 mAP Rank-1 CAN [14] yes 24.”
• Attribute-Aware Attention Model for Fine-grained Representation Learning (2018, methodology): “In [23], attention mechanism was firstly used in person re-identification by integrating a recurrent attention with the siamese model.”
• Learning Deep Context-Aware Features over Body and Latent Parts for Person Re-identification (2017, methodology): “Market1501: For the Market1501 dataset, several stateof-the-art methods are compared, including Bag of Words (BOW) [53], Weighted Approximate Rank Component Analysis (WARCA) [15], Discriminative Null Space (DNS) [47], Spatially Constrained Similarity function on Polynomial feature map (SCSP) [2], an”
• Human Semantic Parsing for Person Re-identification (2018, background): “Taking a slightly different approach, in [26, 30], authors develop attention-based models where respectively, a Long Short-Term Memory (LSTM) [18] and a gradient-based attention model dynamically focus on distinctive regions in the image.”
• Multi-level Factorisation Net for Person Re-identification (2018, methodology): “Deep Neural Networks for Person Re-ID Most recent person Re-ID methods train deep DNN models with various learning objectives including classification, verification and triplet ranking losses [42, 38, 24, 32, 27].”
• Pedestrian Alignment Network for Large-scale Person Re-Identification (2017, methodology): “Similarly, [Liu et al., 2016a] propose a soft attention-based model to focus on parts and combine CNN with LSTM components selectively; its limitation also consists of the computation inefficiency.”
• A Strong Baseline and Batch Normalization Neck for Deep Person Re-Identification (2019, methodology): “The most widely used baseline based on metric learning is training model with triplet loss [14].”
• Deep Representation Learning With Part Loss for Person Re-Identification (2017, background): “[40] propose an end-to-end Comparative”
• Unsupervised Person Re-identification (2017, background): “On the other hand, comparing with deep similarity learning methods [4], [5], [6], the representation learning methods are more extendable for large galleries.”
• Attention mechanisms in computer vision: A survey (2021, neutral): “[8], [9], [15], [21], [31], [32], [34], [35] , [22], [26], [62], [63], [64], [65], [66], [67] , [41], [68], [69], [70], [71], [72], [73], [74] , [8], [34], [42], [43], [75], [76], [77], [78] , [27], [44], [45], [46], [79], [80], [81], [82] , [61], [83], [84], [85], [86], [87], [88], [89] , [47], [90”
• Mancs: A Multi-task Attentional Network with Curriculum Sampling for Person Re-Identification (2018, methodology): “MSCAN [15] uses a spatial transformation network (STN) [13] to get several attention regions and then extracts local feature from the regions.”
• Person Re-identification: Past, Present and Future (2016, result): “Similar to [72], Liu et al. [73] propose integrating a soft attention based model in a siamese network to adaptively focus on the important local parts of an input image pair; however, this method is also limited by computational inefficiency.”
• Dual Attention Matching Network for Context-Aware Feature Sequence Based Person Re-identification (2018, methodology): “To demonstrate the effectiveness of our dual attention mechanism, in Table 3, we compare our method with several existing attentive methods, including CAN [24], HP-Net [25], ST-Dataset Comparison to Other Feature Sequence / Feature Set based Methods.”
• Attention-Aware Compositional Network for Person Re-identification (2018, methodology): “Since the attention mechanism is effective in understanding images, it has been widely used in various tasks, including machine translation [4], visual question answering [46], object detection [3], semantic segmentation [8], pose estimation [13] and person re-identification [30].”