Deep Learning for Person Re-Identification: A Survey and Outlook

Ye, Mang; Shen, Jianbing; Lin, Gaojie; Xiang, Tao; Shao, Ling; Hoi, Steven C. H.

doi:10.1109/tpami.2021.3054775

Cited by 1,242 publications

(548 citation statements)

References 238 publications

Supporting

Mentioning

468

Contrasting

Order By: Relevance

“…Deep learning-based systems have been developed that range from the detection of contextual situations to the efficient recognition of faces [17], [18]. Although features are automatically generated, feature representation learning is an important process [19]. The most recent work in aesthetic identification by Bari, Brandon, and Gavrilova [4] used a custom CNN architecture to analyze user images.…”

Section: Related Workmentioning

confidence: 99%

Person Identification From Audio Aesthetic

Sieu

Gavrilova

2021

IEEE Access

View full text Add to dashboard Cite

Behavioral biometrics survey actions rather than the physical traits of the person. Within this categorization, social behavioral biometrics utilizes an individual's communications for biometric analysis. The investigation of the uniqueness of human preferences and their implications to other aspects of an individual, such as personality or gender, is both a psychological and a biometric problem. An emerging approach is the usage of an individual's aesthetic preferences for the purpose of person identification. Recent research into the identification from visual aesthetics has found that these preferences hold significant discriminatory value. However, aesthetic identification has only been conducted through a visual medium via a set of liked images. The contribution of this work is the development of the first audio aesthetic preference system for person identification. The proposed system extracts descriptive intra-song and intersong features from a set of songs favored by users and utilizes an ensemble of classifiers for prediction. The final decision is optimized by a genetic algorithm. Experimental results demonstrate that the developed audio aesthetic system achieves 95% user recognition accuracy on both proprietary and public audio datasets.INDEX TERMS audio aesthetics, behavioral biometrics, biometric security, human-machine interactions, pattern recognition

show abstract

Section: Related Workmentioning

confidence: 99%

Person Identification From Audio Aesthetic

Sieu

Gavrilova

2021

IEEE Access

View full text Add to dashboard Cite

show abstract

“…Unsupervised person re-identification aims at exploring discriminative information from unlabeled person images without expensive data annotation, which is more suitable for real applications. Benefit from the success of deep learning, deeply unsupervised person re-identification methods are popular in recent years [22]. some works focus on purely unsupervised person re-identification without any external dataset or identity annoation.…”

Section: A Unsupervised Person Re-identificationmentioning

confidence: 99%

Unsupervised Horizontal Pyramid Similarity Learning for Cross-Domain Adaptive Person Re-Identification

Wei

Liu

et al. 2021

IEEE Access

View full text Add to dashboard Cite

Although person re-identification has made great progress, unsupervised cross-domain adaptive person re-identification is still a challenging problem. With no labeled data in target domain, the performance may have a significant drop. In this paper, we propose an unsupervised cross-domain adaptive person re-identification framework based on horizontal pyramid similarity learning (UHPS). Firstly, horizontal pyramid features are extracted by dividing the deep feature maps into different number of partial feature bins. These feature bins with diverse scales can incorporate not only the global information but also local information in different spatial scales, making the framework more robust in complex environment. Then, horizontal pyramid similarity learning is proposed with the mechanism of fusing together the internal similarity of the target domain and the similarity between the source domain and target domain. Finally, the unsupervised clustering algorithm DBSCAN embeded with the horizontal pyramid similarity is employed to select training data in the target domain and estimate the pseudo labels in each training iteration, for the purpose of adapting the framework to the target domain. The results on Market1501 and DukeMTMC-reID confirm that the proposed framework can adapt to the target domain effectively and outperforms the state-of-the-art unsupervised cross domain person re-identification approaches.

show abstract

“…Such framework allows to establish similarities between images by generating concise image descriptions (an embedding space) that can be compared with simple operations (e.g., dot products, distance computations). This progress in the field of computer vision is behind the advances in re-identification techniques, applied to persons (Ye et al, 2021;Zhong et al, 2019) or vehicles (Chu et al, 2019;Zheng et al, 2019), for instance. Such approaches bear many similar traits to the maritime surveillance problem considered here and inspired our analysis.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning for Vessel Detection and Identification From Spaceborne Optical Imagery

Matasci¹,

Plante²,

Kasa³

et al. 2021

ISPRS Ann. Photogramm. Remote Sens. Spatial Inf. Sci.

View full text Add to dashboard Cite

Abstract. We present a deep learning-based vessel detection and (re-)identification approach from spaceborne optical images. We introduce these two components as part of a maritime surveillance from space pipeline and present experimental results on challenging real-world maritime datasets derived from WorldView imagery. First, we developed a vessel detection model based on RetinaNet achieving a performance of 0.795 F1-score on a challenging multi-scale dataset. We then collected a large-scale dataset for vessel identification by applying the detection model on 200+ optical images, detecting the vessels therein and assigning them an identity via an Automatic Identification System association framework. A vessel re-identification model based on Twin neural networks has then been trained on this dataset featuring 2500+ unique vessels with multiple repeated occurrences across different acquisitions. The model allows to naturally establish similarities between vessel images. It returns a relevant ranking of candidate vessels from a database when provided an input image for a specific vessel the user might be interested in, with top-1 and top-10 accuracies of 38.7% and 76.5%, respectively. This study demonstrates the potential offered by the latest advances in deep learning and computer vision when applied to optical remote sensing imagery in a maritime context, opening new opportunities for automated vessel monitoring and tracking capabilities from space.

show abstract

Deep Learning for Person Re-Identification: A Survey and Outlook

Cited by 1,242 publications

References 238 publications

Person Identification From Audio Aesthetic

Person Identification From Audio Aesthetic

Unsupervised Horizontal Pyramid Similarity Learning for Cross-Domain Adaptive Person Re-Identification

Deep Learning for Vessel Detection and Identification From Spaceborne Optical Imagery

Contact Info

Product

Resources

About