Semantic-Guided Pixel Sampling for Cloth-Changing Person Re-Identification

Shu, Xiujun; Li, Ge; Wang, Xiao; Ruan, Weijian; Tian, Qi

doi:10.1109/lsp.2021.3091924

Cited by 46 publications

(21 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Therefore, to accelerate the development of cloth-changing person ReID techniques, some cloth-changing person ReID datasets have been built and released, such as LTCC [22], PRCC [21], Celeb-reID [20], and NKUP [30]. Moreover, several researchers [20], [21], [22], [31], [32], [33] have made some attempts to address this problem and then assessed their performance on a certain dataset. For example, Yang et al [21] proposed a SPT+ASE module, where human contour sketching information was used to substitute for human color information.…”

Section: B Cloth-changing Person Reidmentioning

confidence: 99%

“…Hong et al [31] proposed a fine-grained shapeappearance mutual learning framework that can learn finegrained discriminative body shape knowledge in a shaped stream and transfer it to an appearance stream to complement the clothing-unrelated knowledge in the appearance features. Shu et al [33] proposed a semantic-guided pixel sampling approach for the cloth-changing person re-ID task which forces the model to automatically learn clothing-irrelevant cues that are irrelevant to upper clothes and pants. Gu et al [41] proposed a Clothes-based Adversarial Loss (CAL) to mine clothes irrelevant features from the original RGB images by penalizing the predictive power of the ReID model.…”

Section: B Cloth-changing Person Reidmentioning

confidence: 99%

“…Since the cloth-changing person ReID task is a new and challenging research topic, only a few works have been published, including SPT+ASE (TPAMI2021) [21], SE+CESD (ACCV 2020) [22], ReIDCaps (TCSVT 2020) [20], Pixel Sampling (ISPL 2021) [33], FSAM (CVPR 2021) [31], and CAL [41] (CVPR 2022). Additionally, in the cloth-changing person ReID task, traditional person ReID algorithms are often employed, such as PCB (ECCV 2018) [9], MGN (ACM MM 2018) [13], ResNet50 (CVPR 2016) [39], DenseNet121 (CVPR 2017) [40], and Swin Transformer (ICCV 2021) [37].…”

Section: A Competitorsmentioning

confidence: 99%

“…In addition, the learning rate is decreased by a factor of 0.1 after 40 epochs. Finally, the cumulative matching characteristics (CMCs), rank-1, and mean average precision (mAP) are often utilized as the evaluation metrics in person ReID tasks [21], [15], [20], [33]; thus, we also strictly follow these metrics in our experiments.…”

Section: B Implementation Detailsmentioning

confidence: 99%

See 3 more Smart Citations

A Semantic-aware Attention and Visual Shielding Network for Cloth-changing Person Re-identification

Gao¹,

Wei²,

Guan³

et al. 2022

Preprint

View full text Add to dashboard Cite

Section: B Cloth-changing Person Reidmentioning

confidence: 99%

Section: B Cloth-changing Person Reidmentioning

confidence: 99%

Section: A Competitorsmentioning

confidence: 99%

Section: B Implementation Detailsmentioning

confidence: 99%

See 2 more Smart Citations

A Semantic-aware Attention and Visual Shielding Network for Cloth-changing Person Re-identification

Gao¹,

Wei²,

Guan³

et al. 2022

Preprint

View full text Add to dashboard Cite

“…To address the above practical problem, Cloth-Changing Person Re-identification (CC-ReID) has drawn increasing attention in recent years [5][6][7][8][9][10][11][12][13][14]. Existing CC-ReID methods often focus on mining identity-relevant cues from pre-defined biometric traits.…”

Section: Introductionmentioning

confidence: 99%

Identity-Sensitive Knowledge Propagation for Cloth-Changing Person Re-Identification

Liu

Shi

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Cloth-changing person re-identification (CC-ReID), which aims to match person identities under clothing changes, is a new rising research topic in recent years. However, typical biometrics-based CC-ReID methods often require cumbersome pose or body part estimators to learn cloth-irrelevant features from human biometric traits, which comes with high computational costs. Besides, the performance is significantly limited due to the resolution degradation of surveillance images. To address the above limitations, we propose an effective Identity-Sensitive Knowledge Propagation framework (DeSKPro) for CC-ReID. Specifically, a Cloth-irrelevant Spatial Attention module is introduced to eliminate the distraction of clothing appearance by acquiring knowledge from the human parsing module. To mitigate the resolution degradation issue and mine identity-sensitive cues from human faces, we propose to restore the missing facial details using prior facial knowledge, which is then propagated to a smaller network. After training, the extra computations for human parsing or face restoration are no longer required. Extensive experiments show that our framework outperforms state-ofthe-art methods by a large margin. Our code is available at https://github.com/KimbingNg/DeskPro.

show abstract

A benchmark for clothes variation in person re‐identification

Wang

Chen

et al. 2020

Int J Intell Syst

View full text Add to dashboard Cite

Person re-identification (re-ID) has drawn attention significantly in the computer vision society due to its application and research significance. It aims to retrieve a person of interest across different camera views. However, there are still several factors that hinder the applications of person re-ID. In fact, most common data sets either assume that pedestrians do not change their clothing across different camera views or are taken under constrained environments. Those constraints simplify the person re-ID task and contribute to early development of person re-ID, yet a person has a great possibility to change clothes in real life. To facilitate the research toward conquering those issues, this paper mainly introduces a new benchmark data set for person reidentification. To the best of our knowledge, this data set is currently the most diverse for person re-identification. It contains 107 persons with 9,738 images, captured in 15 indoor/outdoor scenes from September 2019 to December 2019, varying according to viewpoints, lighting, resolutions, human pose, seasons, backgrounds, and clothes especially. We hope that this benchmark data set will encourage further research on person re-identification with clothes variation. Moreover, we also perform extensive analyses on this data set using several

show abstract

Semantic-Guided Pixel Sampling for Cloth-Changing Person Re-Identification

Cited by 46 publications

References 23 publications

A Semantic-aware Attention and Visual Shielding Network for Cloth-changing Person Re-identification

A Semantic-aware Attention and Visual Shielding Network for Cloth-changing Person Re-identification

Identity-Sensitive Knowledge Propagation for Cloth-Changing Person Re-Identification

A benchmark for clothes variation in person re‐identification

Contact Info

Product

Resources

About