Person Re-identification: Past, Present and Future

By Liang Zheng, Yi Yang, Alexander G. Hauptmann. arXiv.org, 2016

1,176 citations recorded by the source corpus.

Person re-identification (re-ID) has become increasingly popular in the community due to its application and research significance. It aims at spotting a person of interest in other cameras. In the early days, hand-crafted algorithms and small-scale evaluation were predominantly reported. Recent years have witnessed the emergence of large-scale datasets and deep learning systems which make use of large data volumes. Considering different tasks, we classify most current re-ID methods into two classes, i.e., image-based and video-based; in both tasks, hand-crafted and deep learning systems will be reviewed. Moreover, two new re-ID tasks which are much closer to real-world applications are described and discussed, i.e., end-to-end re-ID and fast re-ID in very large galleries. This paper: 1) introduces the history of person re-ID and its relationship with image classification and instance retrieval; 2) surveys a broad selection of the hand-crafted systems and the large-scale methods in both image- and video-based re-ID; 3) describes critical future directions in end-to-end re-ID and fast retrieval in large galleries; and 4) finally briefs some important yet under-developed issues.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/367008b91eb57c5ea64ef7520dfcabc0c5c85532

How researchers see this paper

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

In Computer Vision and Pattern Recognition

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

Citations in the corpus

• Perceive Where to Focus: Learning Visibility-Aware Part-Level Features for Partial Person Re-Identification (2019, methodology): “Deep learning methods currently dominate the re-ID research community with significant superiority on retrieval accuracy [32].”
• Deep Transfer Learning for Person Re-Identification (2016, background): “Despite the best efforts from the computer vision researchers, it remains an unsolved problem [1].”
• Pose-Invariant Embedding for Deep Person Re-Identification (2017, background): “It is pointed out in [2] that the Siamese model only uses weak”
• Hierarchical Discriminative Learning for Visible Thermal Person Re-Identification (2018, background): “A detailed survey about visible image based re-identification can be found in (Zheng, Yang, and Hauptmann 2016).”
• Cross-Modality Paired-Images Generation for RGB-Infrared Person Re-Identification (2020, methodology): “Zheng et al. (Zheng, Zheng, and Yang 2017) use a GAN model to generate unlabeled images as data augmentation.”
• RGB-Infrared Cross-Modality Person Re-Identification via Joint Pixel and Feature Alignment (2019, methodology): “Existing methods can be grouped into hand-crafted descriptors [22, 36, 19, 34], metric learning methods [43, 16, 20] and deep learning algorithms [42, 11, 28, 21, 44, 30, 15, 35, 33, 18, 17, 1, 31].”
• Enhancing Person Re-identification in a Self-Trained Subspace (2017, background): “Person Re-identification (re-ID) [1]–[6] aims to recognize an individual across spatially disjoint cameras.”
• Avoiding Overfitting: A Survey on Regularization Methods for Convolutional Neural Networks (2022, methodology): “In this work, the authors evaluated the method on three different classification datasets (CIFAR-10, CIFAR-100, and Fashion-MNIST), the PASCAL VOC 2007 [8] dataset for object detection, and three different CNN architectures for person re-identification (IDE [74], TriNet [19], and SVDNet [50]).”
• Invariance Matters: Exemplar Memory for Domain Adaptive Person Re-Identification (2019, methodology): “Due to the identities of source images are available, we treat the training process of the source domain as a classification problem [38].”
• Pedestrian Alignment Network for Large-scale Person Re-Identification (2017, methodology): “The learned embeddings extracted from the fine-tuned CNNs are shown to outperform the hand-crafted features [Zheng et al., 2016b,Xiao et al., 2016,Zhong et al., 2017].”
• Beyond Part Models: Person Retrieval with Refined Part Pooling (2017, methodology): “The state of the art on most person retrieval datasets is presently maintained by deep learning methods [44].”
• Continual Representation Learning for Biometric Identification (2020, background): “For example, the intelligent security system [65, 77] in a city or an airport captures millions of new images every day.”
• Lifelong Person Re-Identification via Adaptive Knowledge Accumulation (2021, background): “For instance, smart surveillance systems [53, 15] over multiple crossroads capture millions of new images every day, and they are required to have the ability of incremental or lifelong learning.”
• ABD-Net: Attentive but Diverse Person Re-Identification (2019, background): “We refer readers to [33] for a more comprehensive review.”