Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios

doi:10.19682/j.cnki.1005-8885.2022.0007

中国邮电高校学报(英文) ›› 2022, Vol. 29 ›› Issue (5): 21-29.doi: 10.19682/j.cnki.1005-8885.2022.0007

所属专题： Special Topic on Artificial Intelligence of Things

• Special Topic: Artificial Intelligence of Things • 上一篇下一篇

Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios

李娜¹,武阳阳¹,刘颖¹,李大湘²,高嘉乐¹

1. 西安邮电大学
2. 西安邮电学院通信工程学院

收稿日期:2021-05-12 修回日期:2021-09-16 出版日期:2022-10-31 发布日期:2022-10-28
通讯作者: 武阳阳 E-mail:2745027040@qq.com
基金资助:
国家自然科学基金

Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios

Li Na¹,², Wu Yangyang¹, Liu Ying¹,², Li Daxiang¹,², Gao Jiale¹

Received:2021-05-12 Revised:2021-09-16 Online:2022-10-31 Published:2022-10-28

摘要/Abstract

摘要：

Pedestrian attribute recognition is often considered as a multi-label image classification task. In order to make full use of attribute-related location information, a saliency guided sel-attention network ( SGSA-Net) was proposed to weakly supervise attribute localization， without annotations of attribute-related regions. Saliency priors were integrated into the spatial attention module ( SAM ). Meanwhile，channel-wise attention and spatial attention were introduced into the network. Moreover, a weighted binary cross-entropy loss ( WCEL) function was employed to handle the imbalance of training data. Extensive experiments on richly annotated pedestrian ( RAP) and pedestrian attribute ( PETA) datasets demonstrated that SGSA-Net outperformed other state-of-the-art methods.

关键词: pedestrian attribute recognition|saliency detection|self-attention mechanism

Abstract:

Pedestrian attribute recognition is often considered as a multi-label image classification task. In order to make full use of attribute-related location information, a saliency guided sel-attention network ( SGSA-Net) was proposed to weakly supervise attribute localization，without annotations of attribute-related regions. Saliency priors were integrated into the spatial attention module ( SAM ). Meanwhile, channel-wise attention and spatial attention were introduced into the network. Moreover, a weighted binary cross-entropy loss ( WCEL) function was employed to handle the imbalance of training data. Extensive experiments on richly annotated pedestrian ( RAP) and pedestrian attribute ( PETA) datasets demonstrated that SGSA-Net outperformed other state-of-the-art methods.

Key words: pedestrian attribute recognition|saliency detection|self-attention mechanism

Li Na, Wu Yangyang, Liu Ying, Li Daxiang, Gao Jiale. Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios[J]. The Journal of China Universities of Posts and Telecommunications, 2022, 29(5): 21-29.

参考文献

[1] LIN Y T, ZHENG L, ZHENG Z D, et al. Improving person re-identification by attribute and identity learning. Pattern Recognition, 2019, 95: 151 -161.

[2] SU C, YANG F, ZHANG S L, et al. Multi-task learning with low rank attribute embedding for multi-camera person re-identification. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018, 40(5): 1167 -1181.

[3] ZHOU B L, TANG X O, WANG X G. Learning collective crowd behaviors with dynamic pedestrian-agents. International Journal of Computer Vision, 2015, 111(1): 50 -68.

[4] ZHU C, PENG Y X. A boosted multi-task model for pedestrian detection with occlusion handling. IEEE Transactions on Image Processing, 2015, 24(12): 5619 -5629.

[5] LI D W, CHEN X T, HUANG K Q. Multi-attribute learning for pedestrian attribute recognition in surveillance scenarios. Proceedings of the 3rd IAPR Asian Conference on Pattern Recognition ( ACPR’15), 2015, Nov 3 - 6, Kuala Lumpur, Malaysia. Piscataway, NJ, USA: IEEE, 2015: 111 -115.

[6] SUDOWE P, SPITZER H, LEIBE B. Person attribute recognition with a jointly-trained holistic CNN model. Proceedings of the 2015 IEEE International Conference on Computer Vision Workshop (ICCVW’15), 2015, Dec 7 - 13, Santiago, Chile. Piscataway, NJ, USA: IEEE, 2015: 87 -95.

[7] LI D W, CHEN X T, ZHANG Z, et al. Pose guided deep model for pedestrian attribute recognition in surveillance scenarios. Proceedings of the 2018 IEEE International Conference on Multimedia and Expo (ICME’18), 2018, Jul 23 -27, San Diego, CA, USA. Piscataway, NJ, USA: IEEE, 2018: 1 -6.

[8] LIU P Z, LIU X H, YAN J J, et al. Localization guided learning for pedestrian attribute recognition. Proceedings of the 29th British Machine Vision Conference ( BMVC’18), 2018, Sept 3 - 6, Newcastle, UK. Durham, UK: British Machine Vision Association (BMVA), 2018: 1 -11.

[9] WANG J Y, ZHU X T, GONG S G, et al. Attribute recognition by joint recurrent learning of context and correlation. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV’17), 2017, Oct 22 -29, Venice, Italy. Piscataway, NJ, USA: IEEE, 2017: 531 -540.

[10] ZHAO X, SANG L F, DING G G, et al. Grouping attribute recognition for pedestrian with joint recurrent learning. Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI-ECAI’18 ), 2018, Jul 13 - 19, Stockholm, Sweden. 2018: 3177 -3183.

[11] YAMAGUCHI K, KIAPOUR M H, ORTIZ L E, et al. Retrieving similar styles to parse clothing. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015, 37(5): 1028 -1040.

[12] LI A N, LIU L Q, WANG K, et al. Clothing attributes assisted person reidentification. IEEE Transactions on Circuits and Systems for Video Technology, 2015, 25(5): 869 -878.

[13] DENG Y B, LUO P, LOY C C, et al. Pedestrian attribute recognition at far distance. Proceedings of the 22nd ACM International Conference on Multimedia ( MM’14), 2014, Nov 3 -7, Orlando, FL, USA. New York, NY, USA: ACM, 2014: 789 -792.

[14] BEKELE E, LAWSON W. The deeper, the better: analysis of person attributes recognition. Proceedings of the 14th IEEE International Conference on Automatic Face and Gesture Recognition ( FG’19 ), 2019, May 14 - 18, Lille, France. Piscataway, NJ, USA: IEEE, 2019: 1 -8.

[15] LI D W, CHEN X T, ZHANG Z, et al. Learning deep context aware features over body and latent parts for person re- identification. Proceedings of the 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’17), 2017, Jul 21 - 26, Honolulu, HI, USA. Piscataway, NJ, USA: IEEE, 2017: 384 -393.

[16] ZITNICK C L, DOLLAR P. Edge boxes: locating object proposals from edges. Proceedings of the 13th European Conference on Computer Vision ( ECCV’14 ), 2014, Sep 6 - 12, Zurich, Switzerland. LNCS 8693. Berlin, Germany: Springer, 2014:391-405.

[17] ZHOU Y, YU K, LENG B, et al. Weakly-supervised learning of mid-level features for pedestrian attribute recognition and localization. Proceedings of the 28th British Machine Vision Conference ( BMVC’17 ), 2017, Sep 4 - 7, London, UK. Durham, UK: British Machine Vision Association ( BMVA), 2017: 1 -12.

[18] JI Z, HE E L, WANG H R, et al. Image-attribute reciprocally guided attention network for pedestrian attribute recognition. Pattern Recognition Letters, 2019, 120: 89 -95.

[19] TAN Z C, YANG Y, WAN J, et al. Attention-based pedestrian attribute analysis. IEEE Transactions on Image Processing, 2019, 28(12): 6126 -6140.

[20] YAGHOUBI E, BORZA D, NEVES J, et al. An attention-based deep learning model for multiple pedestrian attributes recognition. Image and Vision Computing, 2020, 102: Article 103981.

[21] ZENG Y, ZHUGE Y Z, LU H C, et al. Joint Learning of saliency detection and weakly supervised semantic segmentation. Proceedings of the 2019 IEEE/ CVF International Conference on Computer Vision ( ICCV’19), 2019, Oct 27 - Nov 2, Seoul, Republic of Korea. Piscataway, NJ, USA: IEEE, 2019: 7222 - 7232.

[22] QUISPE R, PEDRINI H. Improved person re-identification based on saliency and semantic parsing with deep neural network models. Image and Vision Computing, 2019, 92: Article 103809.

[23] HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition. Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR’16), 2016, Jun 27 - 30, Las Vegas, NV, USA, Piscataway, NJ, USA: IEEE, 2016: 770 -778.

[24] CHEN X S, LU H H, CHENG K L, et al. Sequentially refined spatial and channel-wise feature aggregation in encoder-decoder network for single image dehazing. Proceedings of the 2019 IEEE International Conference on Image Processing (ICIP’19), 2019, Sep 22 - 25, Taipei, China. Piscataway, NJ, USA: IEEE, 2019: 2776 -2780.

[25] ZHANG H, GOODFELLOW I, METAXAS D, et al. Self- attention generative adversarial networks. Proceedings of the 36th International Conference on Machine Learning (ICML’19), 2019, Jun 10 - 15, Long Beach, CA, USA. Proceedings of Machine Learning Research (PMLR) 97. 2019: 7354 -7363.

[26] LI D W, ZHANG Z, CHEN X T, et al. A richly annotated dataset for pedestrian attribute recognition. arXiv Preprint, arXiv: 1603. 07054, 2016.

[27] LI D W, ZHANG Z, CHEN X T, et al. A richly annotated pedestrian dataset for person retrieval in real surveillance scenarios. IEEE Transactions on Image Processing, 2019, 28(4): 1575 -1590.

[28] LIU X H, ZHAO H Y, TIAN M Q, et al. HydraPlus-Net: attentive deep features for pedestrian analysis. Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV’17), 2017, Oct 22 -29, Venice, Italy. Piscataway, NJ,USA: IEEE, 2017: 350 -359.

[29] SARFRAZ M S, SCHUMANN A, WANG Y, et al. Deep view sensitive pedestrian attribute inference in an end-to-end model.Proceedings of the 28th British Machine Vision Conference(BMVC’17), 2017, Sep 4 - 7, London, UK. Durham, UK:British Machine Vision Association (BMVA), 2017: 1 -13.

Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios

Saliency guided self-attention network for pedestrian attribute recognition in surveillance scenarios

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 0

编辑推荐

Metrics

本文评价