Abstract
Domain Generalizable Person Re-Identification (DG-ReID) is a more practical ReID task that is trained from multiple source domains and tested on the unseen target domains. Most existing methods are challenged for dealing with the shared and specific characteristics among different domains, which is called the domain conflict problem. To address this problem, we present an Adaptive Cross-domain Learning (ACL) framework equipped with a CrOss-Domain Embedding Block (CODE-Block) to maintain a common feature space for capturing both the domain-invariant and the domain-specific features, while dynamically mining the relations across different domains. Moreover, our model adaptively adjusts the architecture to focus on learning the corresponding features of a single domain at a time without interference from the biased features of other domains. Specifically, the CODE-Block is composed of two complementary branches, a dynamic branch for extracting domain-adaptive features and a static branch for extracting the domain-invariant features. Extensive experiments demonstrate that the proposed approach achieves state-of-the-art performances on the popular benchmarks. Under Protocol-2, our method outperforms previous SOTA by 7.8% and 7.6% in terms of mAP and rank-1 accuracy.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Khosla, A., Zhou, T., Malisiewicz, T., Efros, A.A., Torralba, A.: Undoing the damage of dataset bias. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012. LNCS, vol. 7572, pp. 158–171. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-33718-5_12
Boqiang, X., Jian, L., Lingxiao, H., Zhenan, S.: Meta: Mimicking embedding via others’ aggregation for generalizable person re-identification. arXiv preprint arXiv:2112.08684 (2021)
Change, L.C., Tao, X., Shaogang, G.: Time-delayed correlation analysis for multi-camera activity understanding. IJCV 90(1), 106–129 (2010)
Chen, P., et al.: Dual distribution alignment network for generalizable person re-identification. In: AAAI, pp. 1054–1062 (2021)
Luo, C., Song, C., Zhang, Z.: Generalizing person re-identification by camera-aware invariance learning and cross-domain mixup. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12360, pp. 224–241. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58555-6_14
Ci-Siang, L., Yuan-Chia, C., Frank, W.Y.C.: Domain generalized person re-identification via cross-domain episodic learning. In: ICPR, pp. 6758–6763 (2021)
Da, L., Yongxin, Y., Yi-Zhe, S., M, H.T.: Learning to generalize: Meta-learning for domain generalization. In: AAAI, pp. 3490–3497 (2018)
David, H., Andrew, D., V, L.Q.: Hypernetworks. arXiv preprint arXiv:1609.09106 (2016)
Dominic, Z., von Oswald, J., Seijin, K., João, S., Grewe, B.F.: Meta-learning via hypernetworks. In: NeurIPS (2020)
Gray, D., Tao, H.: Viewpoint invariant pedestrian recognition with an ensemble of localized features. In: Forsyth, D., Torr, P., Zisserman, A. (eds.) ECCV 2008. LNCS, vol. 5302, pp. 262–275. Springer, Heidelberg (2008). https://doi.org/10.1007/978-3-540-88682-2_21
Feihu, Z., W, W.B.: Supplementary meta-learning: towards a dynamic model for deep neural networks. In: ICCV, pp. 4344–4353 (2017)
Fengliang, Q., Bo, Y., Leilei, C., Hongbin, W.: Stronger baseline for person re-identification. arXiv preprint arXiv:2112.01059 (2021)
Gao, H., Shichen, L., der Maaten Laurens, V., Q, W.K.: Condensenet: an efficient densenet using learned group convolutions. In: CVPR, pp. 2752–2761 (2018)
Hao, L., et al.: A strong baseline and batch normalization neck for deep person re-identification. TMM 22(10), 2597–2609 (2019)
Haoliang, L., Jialin, P.S., Shiqi, W., C, K.A.: Domain generalization with adversarial feature learning. In: CVPR, pp. 5400–5409 (2018)
Hubara, I., Courbariaux, M., Soudry, D., El-Yaniv, R., Bengio, Y.: Binarized neural networks. In: NeurIPS, pp. 4107–4115 (2016)
Jie, H., Li, S., Gang, S.: Squeeze-and-excitation networks. In: CVPR, pp. 7132–7141 (2018)
Song, J., Yang, Y., Song, Y.Z., Xiang, T., Hospedales, T.M.: Generalizable person re-identification by domain-invariant mapping network. In: CVPR, pp. 719–728 (2019)
Kaiwen, Y., Xinmei, T.: Domain-class correlation decomposition for generalizable person re-identification. arXiv preprint arXiv:2106.15206 (2021)
Kaiyang, Z., Ziwei, L., Yu, Q., Tao, X., Change, L.C.: Domain generalization: a survey. arXiv preprint arXiv:2103.02503 (2021)
Kecheng, Z., Jiawei, L., Wei, W., Liang, L., Zheng-jun, Z.: Calibrated feature decomposition for generalizable person re-identification. arXiv preprint arXiv:2111.13945 (2021)
Lanlan, L., Jia, D.: Dynamic deep neural networks: optimizing accuracy-efficiency trade-offs by selective execution. In: AAAI, pp. 3675–3682 (2018)
Liang, Z., Liyue, S., Lu, T., Shengjin, W., Jingdong, W., Qi, T.: Scalable person re-identification: a benchmark. In: CVPR, pp. 1116–1124 (2015)
Liao, S., Shao, L.: Interpretable and generalizable person re-identification with query-adaptive convolution and temporal lifting. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12356, pp. 456–474. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58621-8_27
Lingxiao, H., et al.: Semi-supervised domain generalizable person re-identification. arXiv preprint arXiv:2108.05045 (2021)
Lingxiao, H., Xingyu, L., Wu, L., Xinchen, L., Peng, C., Tao, M.: Fastreid: a pytorch toolbox for general instance re-identification. arXiv preprint arXiv:2006.02631 (2020)
Longhui, W., Shiliang, Z., Wen, G., Qi, T.: Person transfer gan to bridge domain gap for person re-identification. In: CVPR, pp. 79–88 (2018)
Lu, Y., Lingqiao, L., Yunlong, W., Peng, W., Yanning, Z.: Multi-domain joint training for person re-identification. arXiv preprint arXiv:2201.01983 (2022)
M, C.F., Antonio, D., Silvia, B., Barbara, C., Tatiana, T.: Domain generalization by solving jigsaw puzzles. In: CVPR, pp. 2229–2238 (2019)
Mang, Y., Jianbing, S., Gaojie, L., Tao, X., Ling, S., CH, H.S.: Deep learning for person re-identification: a survey and outlook. arXiv preprint arXiv:2001.04193 (2020)
Maniyar, U., Joseph, K.J., Deshmukh, A.A., Dogan, Ü., Balasubramanian, V.N.: Zero-shot domain generalization. arXiv preprint arXiv:2008.07443 (2020)
Hirzer, M., Beleznai, C., Roth, P.M., Bischof, H.: Person re-identification by descriptive and discriminative classification. In: Heyden, A., Kahl, F. (eds.) SCIA 2011. LNCS, vol. 6688, pp. 91–102. Springer, Heidelberg (2011). https://doi.org/10.1007/978-3-642-21227-7_9
Misha, D., Babak, S., Laurent, D., Marc’Aurelio, R., Nando, D.F.: Predicting parameters in deep learning. In: NeurIPS, pp. 2148–2156 (2013)
Muhammad, G., Bastiaan, K.W., Mengjie, Z., David, B.: Domain generalization for object recognition with multi-task autoencoders. In: ICCV, pp. 2551–2559 (2015)
Ma, N., Zhang, X., Huang, J., Sun, J.: WeightNet: revisiting the design space of weight networks. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12360, pp. 776–792. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58555-6_46
Riccardo, V., Vittorio, M.: Addressing model vulnerability to distributional shifts over image transformation sets. In: ICCV, pp. 7980–7989 (2019)
Seokeon, C., Taekyung, K., Minki, J., Hyoungseob, P., Changick, K.: Meta batch-instance normalization for generalizable person re-identification. In: CVPR, pp. 3425–3435 (2021)
Lin, S., Li, C.T., Kot, A.C.: Multi-domain adversarial feature generalization for person re-identification. TIP 30, 1596–1607 (2020)
Shankar, S., Piratla, V., Chakrabarti, S., Chaudhuri, S., Jyothi, P., Sarawagi, S.: Generalizing across domains via cross-gradient training. In: ICLR (2018)
Shengcai, L., Ling, S.: Graph sampling based deep metric learning for generalizable person re-identification. arXiv preprint arXiv:2104.01546 (2021)
Shengcai, L., Ling, S.: Transmatcher: deep image matching through transformers for generalizable person re-identification. In: NeurIPS (2021)
Shijie, Y., et al.: Multiple domain experts collaborative learning: multi-source domain generalization for person re-identification. arXiv preprint arXiv:2105.12355 (2021)
Shiyu, X., Shiliang, Z.: Intra-inter camera similarity for unsupervised person re-identification. In: CVPR, pp. 11926–11935 (2021)
Wang, S., Yu, L., Li, C., Fu, C.-W., Heng, P.-A.: Learning from extrinsic and intrinsic supervisions for domain generalization. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12354, pp. 159–176. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58545-7_10
Tong, X., Hongsheng, L., Wanli, O., Xiaogang, W.: Learning deep feature representations with domain guided dropout for person re-identification. In: CVPR, pp. 1249–1258 (2016)
Tong, X., Shuang, L., Bochao, W., Liang, L., Xiaogang, W.: End-to-end deep learning for person search. arXiv preprint arXiv:1604.01850 (2016)
Wei, L., Rui, Z., Tong, X., Xiaogang, W.: Deepreid: deep filter pairing neural network for person re-identification. In: CVPR, pp. 152–159 (2014)
Wei, L., Xiaogang, W.: Locally aligned feature transforms across views. In: CVPR, pp. 3594–3601 (2013)
Wei-Shi, Z., Shaogang, G., Tao, X.: Associating groups of people. In: BMVC, pp. 1–11 (2009)
Xin, J., Cuiling, L., Wenjun, Z., Zhibo, C., Li, Z.: Style normalization and restitution for generalizable person re-identification. In: CVPR, pp. 3143–3152 (2020)
Pan, X., Luo, P., Shi, J., Tang, X.: Two at once: enhancing learning and generalization capacities via IBN-Net. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11208, pp. 484–500. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01225-0_29
Jia, X., De Brabandere, B., Tuytelaars, T., Gool, L.V.: Dynamic filter networks, pp. 667–675 (2016)
Xun, H., Serge, B.: Arbitrary style transfer in real-time with adaptive instance normalization. In: Proceedings of the IEEE International Conference on Computer Vision, pp. 1501–1510 (2017)
Ya, L., Mingming, G., Xinmei, T., Tongliang, L., Dacheng, T.: Domain generalization via conditional invariant representations. In: AAAI, pp. 3579–3587 (2018)
Li, Y., Tian, X., Gong, M., Liu, Y., Liu, T., Zhang, K., Tao, D.: Deep domain generalization via conditional invariant adversarial networks. In: Ferrari, V., Hebert, M., Sminchisescu, C., Weiss, Y. (eds.) ECCV 2018. LNCS, vol. 11219, pp. 647–663. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-01267-0_38
Yan, B., et al.: Person30k: a dual-meta generalization network for person re-identification. In: CVPR, pp. 2123–2132 (2021)
Yi-Fan, Z., et al.: Learning domain invariant representations for generalizable person re-identification. arXiv preprint arXiv:2103.15890 (2021)
Yizeng, H., Gao, H., Shiji, S., Le, Y., Honghui, W., Yulin, W.: Dynamic neural networks: a survey. TPAMI (2021)
Yongxing, D., Xiaotong, L., Jun, L., Zekun, T., Ling-Yu, D.: Generalizable person re-identification with relevance-aware mixture of experts. In: CVPR, pp. 16145–16154 (2021)
Yuyang, Z., et al.: Learning to generalize unseen domains via memory-based multi-source meta-learning for person re-identification. In: CVPR, pp. 6277–6286 (2021)
Zhang, E., et al.: One for more: Selecting generalizable samples for generalizable reid model. In: AAAI, pp. 3324–3332 (2021)
Zhedong, Z., Liang, Z., Yi, Y.: Unlabeled samples generated by gan improve the person re-identification baseline in vitro. In: ICCV, pp. 3754–3762 (2017)
Zhou, K., Yang, Y., Qiao, Y., Xiang, T.: Domain adaptive ensemble learning. TIP 30, 8008–8018 (2021)
Zhou, K., Yang, Y., Qiao, Y., Xiang, T.: Domain generalization with mixstyle. In: ICLR (2021)
Zhuang, L., Jianguo, L., Zhiqiang, S., Gao, H., Shoumeng, Y., Changshui, Z.: Learning efficient convolutional networks through network slimming. In: ICCV, pp. 2736–2744 (2017)
Zhun, Z., Liang, Z., Donglin, C., Shaozi, L.: Re-ranking person re-identification with k-reciprocal encoding. In: CVPR, pp. 1318–1327 (2017)
Acknowledgement
This work is supported in part by Zhejiang Provincial Natural Science Foundation of China under Grant LR19F020004, National Key Research and Development Program of China under Grant 2020AAA0107400, National Natural Science Foundation of China under Grant U20A20222, Key R &D Program of Zhejiang Province, China (2021C01119), the National Natural Science Foundation of China under Grant (62002320, U19B2043).
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
1 Electronic supplementary material
Below is the link to the electronic supplementary material.
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Zhang, P., Dou, H., Yu, Y., Li, X. (2022). Adaptive Cross-domain Learning for Generalizable Person Re-identification. In: Avidan, S., Brostow, G., Cissé, M., Farinella, G.M., Hassner, T. (eds) Computer Vision – ECCV 2022. ECCV 2022. Lecture Notes in Computer Science, vol 13674. Springer, Cham. https://doi.org/10.1007/978-3-031-19781-9_13
Download citation
DOI: https://doi.org/10.1007/978-3-031-19781-9_13
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-19780-2
Online ISBN: 978-3-031-19781-9
eBook Packages: Computer ScienceComputer Science (R0)