GaitSCM: Causal representation learning for gait recognition
Gait recognition is a promising biometric technology that aims to identify the target subject via walking pattern. Most existing appearance-based methods focus on learning discriminative spatio-temporal representations from gait silhouettes. However, these methods pay less attention to probing the c...
Uložené v:
| Vydané v: | Computer vision and image understanding Ročník 243; s. 103995 |
|---|---|
| Hlavní autori: | , , , , |
| Médium: | Journal Article |
| Jazyk: | English |
| Vydavateľské údaje: |
Elsevier Inc
01.06.2024
|
| Predmet: | |
| ISSN: | 1077-3142, 1090-235X |
| On-line prístup: | Získať plný text |
| Tagy: |
Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
|
| Abstract | Gait recognition is a promising biometric technology that aims to identify the target subject via walking pattern. Most existing appearance-based methods focus on learning discriminative spatio-temporal representations from gait silhouettes. However, these methods pay less attention to probing the causality between identity factors and identity labels, which often mislead the model to learn gait representations that are susceptible to identity-irrelevant factors. In this paper, we attribute the cause that leads to the decline of model generalization under different external conditions to identity-irrelevant factors. We formulate the causalities among the identity factors, identity-irrelevant factors, and identity labels as a structural causal model (SCM). We accordingly propose a novel gait recognition framework named GaitSCM to learn covariate invariant gait representations, which is mainly composed of three components, including feature extraction module, feature disentanglement module, and backdoor adjustment. Specifically, we design a feature extractor with regard to the movement patterns of different body parts to learn fine-grained gait motion features, and then present a two-branch feature decoupling module to disentangle identity features and identity-irrelevant features with the aid of the classification confusion loss. To relieve the negative effect of identity-irrelevant factors, we develop a backdoor adjustment strategy to eliminate spurious associations between identity and identity-irrelevant features, which further facilitates the proposed framework to generate more powerful identity representations. Extensive experiments conducted on two public datasets validate the effectiveness of our method. The average Rank-1 can reach 93.2% and 90.4% on CASIA-B and OU-MVLP datasets, respectively, which verifies the superiority of GaitSCM. Source code is released at: https://github.com/HuoweiCode/GaitSCM.
•A novel causal representation learning framework GaitSCM for gait recognition is proposed.•A causal analysis based on SCM reveals the spurious relation in gait recognition.•Feature decoupling and backdoor adjustment are used to perform causal intervention.•GaitSCM achieves competitive results on the CASIA-B and OU-MVLP datasets. |
|---|---|
| AbstractList | Gait recognition is a promising biometric technology that aims to identify the target subject via walking pattern. Most existing appearance-based methods focus on learning discriminative spatio-temporal representations from gait silhouettes. However, these methods pay less attention to probing the causality between identity factors and identity labels, which often mislead the model to learn gait representations that are susceptible to identity-irrelevant factors. In this paper, we attribute the cause that leads to the decline of model generalization under different external conditions to identity-irrelevant factors. We formulate the causalities among the identity factors, identity-irrelevant factors, and identity labels as a structural causal model (SCM). We accordingly propose a novel gait recognition framework named GaitSCM to learn covariate invariant gait representations, which is mainly composed of three components, including feature extraction module, feature disentanglement module, and backdoor adjustment. Specifically, we design a feature extractor with regard to the movement patterns of different body parts to learn fine-grained gait motion features, and then present a two-branch feature decoupling module to disentangle identity features and identity-irrelevant features with the aid of the classification confusion loss. To relieve the negative effect of identity-irrelevant factors, we develop a backdoor adjustment strategy to eliminate spurious associations between identity and identity-irrelevant features, which further facilitates the proposed framework to generate more powerful identity representations. Extensive experiments conducted on two public datasets validate the effectiveness of our method. The average Rank-1 can reach 93.2% and 90.4% on CASIA-B and OU-MVLP datasets, respectively, which verifies the superiority of GaitSCM. Source code is released at: https://github.com/HuoweiCode/GaitSCM.
•A novel causal representation learning framework GaitSCM for gait recognition is proposed.•A causal analysis based on SCM reveals the spurious relation in gait recognition.•Feature decoupling and backdoor adjustment are used to perform causal intervention.•GaitSCM achieves competitive results on the CASIA-B and OU-MVLP datasets. |
| ArticleNumber | 103995 |
| Author | Tang, Jun Liang, Dong Wang, Nian Huo, Wei Wang, Ke |
| Author_xml | – sequence: 1 givenname: Wei orcidid: 0000-0002-6562-6018 surname: Huo fullname: Huo, Wei organization: School of Electronic and Information Engineering, Anhui University, Hefei 230601, China – sequence: 2 givenname: Ke orcidid: 0000-0001-5083-7552 surname: Wang fullname: Wang, Ke organization: School of Internet, Anhui University, Hefei 230039, China – sequence: 3 givenname: Jun orcidid: 0000-0001-8564-6510 surname: Tang fullname: Tang, Jun email: tangjun@ahu.edu.cn organization: School of Electronic and Information Engineering, Anhui University, Hefei 230601, China – sequence: 4 givenname: Nian orcidid: 0000-0002-3731-5626 surname: Wang fullname: Wang, Nian organization: School of Electronic and Information Engineering, Anhui University, Hefei 230601, China – sequence: 5 givenname: Dong surname: Liang fullname: Liang, Dong organization: School of Internet, Anhui University, Hefei 230039, China |
| BookMark | eNp9kM9KAzEQxoNUsK2-gKd9ga2TZP9FepFFq1DxoIK3kM1OSsqaLcm24NubZT156GGYYWZ-w3zfgsxc75CQWworCrS426_0yR5XDFgWG1yI_ILMKQhIGc-_ZmNdlimnGbsiixD2AJRmgs7JeqPs8F6_3ie1OgbVJR4PHgO6QQ22d0mHyjvrdonpfbKLu3FB9ztnx-k1uTSqC3jzl5fk8-nxo35Ot2-bl_phm2oOMKRNZlRFuYAYORYIxghQZdGWRjCa09JkrMmgaascDK-qQje5blExjShMAXxJqumu9n0IHo3Udvpv8Mp2koIcXZB7ObogRxfk5EJE2T_04O238j_nofUEYRR1suhl0BadxtZG9YNse3sO_wX9wHi3 |
| CitedBy_id | crossref_primary_10_1016_j_cviu_2025_104463 crossref_primary_10_1109_ACCESS_2024_3513541 crossref_primary_10_1007_s11760_024_03765_2 |
| Cites_doi | 10.1109/CVPR52729.2023.00936 10.1109/ICCV48922.2021.01438 10.1016/j.patcog.2021.108453 10.1109/CVPR52688.2022.01138 10.1609/aaai.v36i1.19983 10.1109/CVPR46437.2021.00898 10.1109/CVPR52688.2022.00684 10.1145/3503161.3547936 10.1016/j.ins.2023.03.145 10.1109/TPAMI.2022.3223784 10.1109/TCSVT.2003.821972 10.1016/j.patrec.2019.05.012 10.1109/CVPR42600.2020.01332 10.1109/CVPRW56347.2022.00163 10.1609/aaai.v33i01.33018126 10.1109/TIP.2021.3055936 10.1109/TPAMI.2006.38 10.1109/CVPRW.2017.80 10.1109/CVPR52688.2022.01751 10.1109/TPAMI.2020.2998790 10.1609/aaai.v35i15.17630 10.1016/j.patcog.2019.107069 10.1109/TIFS.2018.2844819 10.1109/CVPR52688.2022.01961 10.1609/aaai.v35i13.17437 10.1145/3394171.3413861 10.1109/CVPR.2019.00483 10.1109/TIP.2022.3229621 10.1109/CVPR42600.2020.01423 10.1016/j.patcog.2003.09.012 10.1109/TCSVT.2023.3241651 10.1109/TCSVT.2022.3175959 10.1109/ICCV48922.2021.01267 10.1109/TCSVT.2020.2975671 10.1109/CVPR.2019.00484 10.1109/ICCVW54120.2021.00456 10.1109/TPAMI.2016.2545669 10.1017/S0266466603004109 10.1609/aaai.v36i3.20271 |
| ContentType | Journal Article |
| Copyright | 2024 Elsevier Inc. |
| Copyright_xml | – notice: 2024 Elsevier Inc. |
| DBID | AAYXX CITATION |
| DOI | 10.1016/j.cviu.2024.103995 |
| DatabaseName | CrossRef |
| DatabaseTitle | CrossRef |
| DatabaseTitleList | |
| DeliveryMethod | fulltext_linktorsrc |
| Discipline | Applied Sciences Engineering Computer Science |
| EISSN | 1090-235X |
| ExternalDocumentID | 10_1016_j_cviu_2024_103995 S1077314224000766 |
| GroupedDBID | --K --M -~X .DC .~1 0R~ 1B1 1~. 1~5 29F 4.4 457 4G. 5GY 5VS 6TJ 7-5 71M 8P~ AABNK AACTN AAEDT AAEDW AAIAV AAIKC AAIKJ AAKOC AALRI AAMNW AAOAW AAQFI AAQXK AAXUO AAYFN ABBOA ABEFU ABJNI ABMAC ABXDB ABYKQ ACDAQ ACGFS ACNNM ACRLP ACZNC ADBBV ADEZE ADFGL ADJOM ADMUD ADTZH AEBSH AECPX AEKER AENEX AFKWA AFTJW AGHFR AGUBO AGYEJ AHJVU AHZHX AIALX AIEXJ AIKHN AITUG AJOXV AKRWK ALMA_UNASSIGNED_HOLDINGS AMFUW AMRAJ AOUOD ASPBG AVWKF AXJTR AZFZN BJAXD BKOJK BLXMC CAG COF CS3 DM4 DU5 EBS EFBJH EJD EO8 EO9 EP2 EP3 F0J F5P FDB FEDTE FGOYB FIRID FNPLU FYGXN G-Q GBLVA GBOLZ HF~ HVGLF HZ~ IHE J1W JJJVA KOM LG5 M41 MO0 N9A O-L O9- OAUVE OZT P-8 P-9 P2P PC. Q38 R2- RIG RNS ROL RPZ SDF SDG SDP SES SEW SPC SPCBC SST SSV SSZ T5K TN5 XPP ZMT ~G- 9DU AATTM AAXKI AAYWO AAYXX ABFNM ABWVN ACLOT ACRPL ACVFH ADCNI ADNMO AEIPS AEUPX AFJKZ AFPUW AGQPQ AIGII AIIUN AKBMS AKYEP ANKPU APXCP CITATION EFKBS EFLBG ~HD |
| ID | FETCH-LOGICAL-c300t-b4fa813901395e6e0ff90a76d7f921517f42b40bd850f3886cb5cdea2cee9f603 |
| ISICitedReferencesCount | 3 |
| ISICitedReferencesURI | http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001223459200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| ISSN | 1077-3142 |
| IngestDate | Tue Nov 18 21:02:58 EST 2025 Sat Nov 29 06:43:57 EST 2025 Sat May 04 15:43:16 EDT 2024 |
| IsPeerReviewed | true |
| IsScholarly | true |
| Keywords | Causal representation learning Global and local feature extractor Disentangled representation learning Gait recognition |
| Language | English |
| LinkModel | OpenURL |
| MergedId | FETCHMERGED-LOGICAL-c300t-b4fa813901395e6e0ff90a76d7f921517f42b40bd850f3886cb5cdea2cee9f603 |
| ORCID | 0000-0001-8564-6510 0000-0001-5083-7552 0000-0002-3731-5626 0000-0002-6562-6018 |
| ParticipantIDs | crossref_citationtrail_10_1016_j_cviu_2024_103995 crossref_primary_10_1016_j_cviu_2024_103995 elsevier_sciencedirect_doi_10_1016_j_cviu_2024_103995 |
| PublicationCentury | 2000 |
| PublicationDate | June 2024 2024-06-00 |
| PublicationDateYYYYMMDD | 2024-06-01 |
| PublicationDate_xml | – month: 06 year: 2024 text: June 2024 |
| PublicationDecade | 2020 |
| PublicationTitle | Computer vision and image understanding |
| PublicationYear | 2024 |
| Publisher | Elsevier Inc |
| Publisher_xml | – name: Elsevier Inc |
| References | Van der Maaten, Hinton (b26) 2008; 9 Wang, Ning, Tan, Hu (b35) 2004; 14 Chen, Wang, Zheng, Zeng, Zou, Cui (b5) 2023; 636 Takemura, Makihara, Muramatsu, Echigo, Yagi (b31) 2018; 10 Xu, Liu, Sun, Yang, Wang, Wang, Fu (b38) 2022; 45 Yang, Zhang, Cai (b41) 2021; 45 Wu, Huang, Wang, Wang, Tan (b37) 2016; 39 Yue, Zhang, Sun, Hua (b45) 2020; 33 Liu, B., Wang, D., Yang, X., Zhou, Y., Yao, R., Shao, Z., Zhao, J., 2022. Show, deconfound and tell: Image captioning with causal inference. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18041–18050. Yu, S., Chen, H., Garcia Reyes, E.B., Poh, N., 2017. Gaitgan: Invariant gait feature extraction using generative adversarial networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. pp. 30–37. He, Zhang, Shan, Wang (b12) 2018; 14 Lin, B., Zhang, S., Bao, F., 2020. Gait recognition with multiple-temporal-scale 3d convolutional neural network. In: Proceedings of the 28th ACM International Conference on Multimedia. pp. 3054–3062. Li, X., Makihara, Y., Xu, C., Yagi, Y., Ren, M., 2020. Gait recognition via semi-supervised disentangled representation learning to identity and covariate features. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 13309–13319. Chen, Luo, Weng, Luo, Li, Tian (b3) 2021; 30 Zhu, B., Niu, Y., Hua, X.-S., Zhang, H., 2022. Cross-domain empirical risk minimization for unbiased long-tailed classification. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 36, pp. 3589–3597. Chen, Z., Tian, Z., Zhu, J., Li, C., Du, S., 2022b. C-CAM: Causal CAM for Weakly Supervised Semantic Segmentation on Medical Image. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11676–11685. Teepe, T., Gilg, J., Herzog, F., Hörmann, S., Rigoll, G., 2022. Towards a deeper understanding of skeleton-based gait recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 1569–1577. Huang, Ben, Gong, Zhang, Yan, Wu (b15) 2022; 32 Han, Bhanu (b11) 2005; 28 Liao, Yu, An, Huang (b22) 2020; 98 Wang, Gao, Xu (b34) 2022; 45 Glymour, Pearl, Jewell (b10) 2016 Fan, C., Peng, Y., Cao, C., Liu, X., Hou, S., Chi, J., Huang, Y., Li, Q., He, Z., 2020. Gaitpart: Temporal part-based model for gait recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 14225–14233. Fan, C., Liang, J., Shen, C., Hou, S., Huang, Y., Yu, S., 2023. OpenGait: Revisiting Gait Recognition Towards Better Practicality. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 9707–9716. Wang, Yang, Wang (b36) 2021; 34 Zhang, Tran, Liu, Liu (b47) 2020; 44 Chao, H., He, Y., Zhang, J., Feng, J., 2019. Gaitset: Regarding gait as a set for cross-view gait recognition. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 33, pp. 8126–8133. Chen, Y., Yang, X., Cham, T.-J., Cai, J., 2022a. Towards unbiased visual emotion recognition via causal intervention. In: Proceedings of the 30th ACM International Conference on Multimedia. pp. 60–69. Zhang, Z., Tran, L., Yin, X., Atoum, Y., Liu, X., Wan, J., Wang, N., 2019b. Gait recognition via disentangled representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4710–4719. Zhang, S., Wang, Y., Li, A., 2021. Cross-view gait recognition with deep universal linear embeddings. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 9095–9104. Seitzer, Schölkopf, Martius (b29) 2021; 34 Yao, Kusakunniran, Wu, Zhang, Tang, Yang (b42) 2021; 150 Xu, Makihara, Li, Yagi, Lu (b39) 2020; 31 Li, B., Han, Z., Li, H., Fu, H., Zhang, C., 2022a. Trustworthy long-tailed classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6970–6979. Hu, Y., Wu, Y., Zhang, L., Wu, X., 2021. A generative adversarial framework for bounding confounded causal effects. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35, pp. 12104–12112. Huang, J., Qin, Y., Qi, J., Sun, Q., Zhang, H., 2022a. Deconfounded visual grounding. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 36, pp. 998–1006. Hermans, Beyer, Leibe (b13) 2017 Huang, X., Zhu, D., Wang, H., Wang, X., Yang, B., He, B., Liu, W., Feng, B., 2021. Context-sensitive temporal feature learning for gait recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 12909–12918. Lin, B., Zhang, S., Yu, X., 2021. Gait recognition via effective global-local feature representation and local temporal aggregation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 14648–14656. Zhang, K., Luo, W., Ma, L., Liu, W., Li, H., 2019a. Learning joint gait representation via quintuplet loss minimization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4700–4709. Zhang, Zhang, Li, Jia, Wang, Tan (b50) 2022; 32 Li, Qiu, Zhao, Zhan, Chen, Wei, Huang (b21) 2022; 124 Zhang, Zhang, Tang, Hua, Sun (b51) 2020; 33 Chai, T., Li, A., Zhang, S., Li, Z., Wang, Y., 2022. Lagrange motion analysis and view embeddings for improved gait recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 20249–20258. Tang, Tao, Qi, Liu, Zhang (b32) 2022 Yam, Nixon, Carter (b40) 2004; 37 Neuberg (b27) 2003; 19 Pearl (b28) 2000 Dang, Luo, Jia, Yan, Chang, Zheng (b7) 2023; 33 Yu, Tan, Tan (b44) 2006; Vol. 4 Staliūnaitė, I., Gorinski, P.J., Iacobacci, I., 2021. Improving commonsense causal reasoning by adversarial training and data augmentation. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35, pp. 13834–13842. Li, X., Makihara, Y., Xu, C., Yagi, Y., 2021. End-to-end model-based gait recognition using synchronized multi-view pose constraint. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 4106–4115. 10.1016/j.cviu.2024.103995_b24 Van der Maaten (10.1016/j.cviu.2024.103995_b26) 2008; 9 10.1016/j.cviu.2024.103995_b23 Xu (10.1016/j.cviu.2024.103995_b38) 2022; 45 10.1016/j.cviu.2024.103995_b25 10.1016/j.cviu.2024.103995_b20 Han (10.1016/j.cviu.2024.103995_b11) 2005; 28 Takemura (10.1016/j.cviu.2024.103995_b31) 2018; 10 Seitzer (10.1016/j.cviu.2024.103995_b29) 2021; 34 Yu (10.1016/j.cviu.2024.103995_b44) 2006; Vol. 4 Yue (10.1016/j.cviu.2024.103995_b45) 2020; 33 10.1016/j.cviu.2024.103995_b1 10.1016/j.cviu.2024.103995_b2 Wu (10.1016/j.cviu.2024.103995_b37) 2016; 39 10.1016/j.cviu.2024.103995_b4 Hermans (10.1016/j.cviu.2024.103995_b13) 2017 Wang (10.1016/j.cviu.2024.103995_b36) 2021; 34 10.1016/j.cviu.2024.103995_b6 10.1016/j.cviu.2024.103995_b8 10.1016/j.cviu.2024.103995_b9 Glymour (10.1016/j.cviu.2024.103995_b10) 2016 10.1016/j.cviu.2024.103995_b30 10.1016/j.cviu.2024.103995_b33 Wang (10.1016/j.cviu.2024.103995_b35) 2004; 14 Pearl (10.1016/j.cviu.2024.103995_b28) 2000 Zhang (10.1016/j.cviu.2024.103995_b51) 2020; 33 Xu (10.1016/j.cviu.2024.103995_b39) 2020; 31 Zhang (10.1016/j.cviu.2024.103995_b47) 2020; 44 Zhang (10.1016/j.cviu.2024.103995_b50) 2022; 32 He (10.1016/j.cviu.2024.103995_b12) 2018; 14 10.1016/j.cviu.2024.103995_b46 Neuberg (10.1016/j.cviu.2024.103995_b27) 2003; 19 10.1016/j.cviu.2024.103995_b48 Yang (10.1016/j.cviu.2024.103995_b41) 2021; 45 10.1016/j.cviu.2024.103995_b43 Huang (10.1016/j.cviu.2024.103995_b15) 2022; 32 10.1016/j.cviu.2024.103995_b49 Chen (10.1016/j.cviu.2024.103995_b3) 2021; 30 Dang (10.1016/j.cviu.2024.103995_b7) 2023; 33 Yam (10.1016/j.cviu.2024.103995_b40) 2004; 37 10.1016/j.cviu.2024.103995_b14 10.1016/j.cviu.2024.103995_b52 Chen (10.1016/j.cviu.2024.103995_b5) 2023; 636 10.1016/j.cviu.2024.103995_b17 10.1016/j.cviu.2024.103995_b16 10.1016/j.cviu.2024.103995_b19 10.1016/j.cviu.2024.103995_b18 Tang (10.1016/j.cviu.2024.103995_b32) 2022 Wang (10.1016/j.cviu.2024.103995_b34) 2022; 45 Liao (10.1016/j.cviu.2024.103995_b22) 2020; 98 Yao (10.1016/j.cviu.2024.103995_b42) 2021; 150 Li (10.1016/j.cviu.2024.103995_b21) 2022; 124 |
| References_xml | – volume: 45 start-page: 7639 year: 2022 end-page: 7653 ident: b38 article-title: PatchMix augmentation to identify causal features in few-shot learning publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 98 year: 2020 ident: b22 article-title: A model-based gait recognition method with body pose and human prior knowledge publication-title: Pattern Recognit. – volume: 636 year: 2023 ident: b5 article-title: GaitAMR: Cross-view gait recognition via aggregated multi-feature representation publication-title: Inform. Sci. – volume: 45 start-page: 3933 year: 2022 end-page: 3948 ident: b34 article-title: Weakly-supervised video object grounding via causal intervention publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – reference: Yu, S., Chen, H., Garcia Reyes, E.B., Poh, N., 2017. Gaitgan: Invariant gait feature extraction using generative adversarial networks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops. pp. 30–37. – reference: Li, X., Makihara, Y., Xu, C., Yagi, Y., 2021. End-to-end model-based gait recognition using synchronized multi-view pose constraint. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 4106–4115. – reference: Chai, T., Li, A., Zhang, S., Li, Z., Wang, Y., 2022. Lagrange motion analysis and view embeddings for improved gait recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 20249–20258. – volume: 33 start-page: 655 year: 2020 end-page: 666 ident: b51 article-title: Causal intervention for weakly-supervised semantic segmentation publication-title: Adv. Neural Inf. Process. Syst. – reference: Lin, B., Zhang, S., Yu, X., 2021. Gait recognition via effective global-local feature representation and local temporal aggregation. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 14648–14656. – reference: Fan, C., Peng, Y., Cao, C., Liu, X., Hou, S., Chi, J., Huang, Y., Li, Q., He, Z., 2020. Gaitpart: Temporal part-based model for gait recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 14225–14233. – reference: Huang, X., Zhu, D., Wang, H., Wang, X., Yang, B., He, B., Liu, W., Feng, B., 2021. Context-sensitive temporal feature learning for gait recognition. In: Proceedings of the IEEE/CVF International Conference on Computer Vision. pp. 12909–12918. – volume: 30 start-page: 3041 year: 2021 end-page: 3055 ident: b3 article-title: Multi-view gait image generation for cross-view gait recognition publication-title: IEEE Trans. Image Process. – volume: 32 start-page: 6967 year: 2022 end-page: 6980 ident: b15 article-title: Enhanced spatial-temporal salience for cross-view gait recognition publication-title: IEEE Trans. Circuits Syst. Video Technol. – start-page: 709 year: 2022 end-page: 726 ident: b32 article-title: Invariant feature learning for generalized long-tailed classification publication-title: Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXIV – volume: 37 start-page: 1057 year: 2004 end-page: 1072 ident: b40 article-title: Automated person recognition by walking and running via model-based approaches publication-title: Pattern Recognit. – volume: 34 start-page: 22905 year: 2021 end-page: 22918 ident: b29 article-title: Causal influence detection for improving efficiency in reinforcement learning publication-title: Adv. Neural Inf. Process. Syst. – reference: Teepe, T., Gilg, J., Herzog, F., Hörmann, S., Rigoll, G., 2022. Towards a deeper understanding of skeleton-based gait recognition. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 1569–1577. – volume: 44 start-page: 345 year: 2020 end-page: 360 ident: b47 article-title: On learning disentangled representations for gait recognition publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – reference: Huang, J., Qin, Y., Qi, J., Sun, Q., Zhang, H., 2022a. Deconfounded visual grounding. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 36, pp. 998–1006. – reference: Zhang, K., Luo, W., Ma, L., Liu, W., Li, H., 2019a. Learning joint gait representation via quintuplet loss minimization. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4700–4709. – reference: Zhang, S., Wang, Y., Li, A., 2021. Cross-view gait recognition with deep universal linear embeddings. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 9095–9104. – reference: Fan, C., Liang, J., Shen, C., Hou, S., Huang, Y., Yu, S., 2023. OpenGait: Revisiting Gait Recognition Towards Better Practicality. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 9707–9716. – year: 2016 ident: b10 article-title: Causal Inference in Statistics: A Primer – volume: 28 start-page: 316 year: 2005 end-page: 322 ident: b11 article-title: Individual recognition using gait energy image publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – reference: Chao, H., He, Y., Zhang, J., Feng, J., 2019. Gaitset: Regarding gait as a set for cross-view gait recognition. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 33, pp. 8126–8133. – reference: Chen, Y., Yang, X., Cham, T.-J., Cai, J., 2022a. Towards unbiased visual emotion recognition via causal intervention. In: Proceedings of the 30th ACM International Conference on Multimedia. pp. 60–69. – reference: Li, X., Makihara, Y., Xu, C., Yagi, Y., Ren, M., 2020. Gait recognition via semi-supervised disentangled representation learning to identity and covariate features. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 13309–13319. – volume: 32 start-page: 509 year: 2022 end-page: 523 ident: b50 article-title: Learning domain invariant representations for generalizable person re-identification publication-title: IEEE Trans. Image Process. – volume: 34 start-page: 21164 year: 2021 end-page: 21175 ident: b36 article-title: Provably efficient causal reinforcement learning with confounded observational data publication-title: Adv. Neural Inf. Process. Syst. – volume: 19 start-page: 675 year: 2003 end-page: 685 ident: b27 article-title: Causality: models, reasoning, and inference, by judea pearl, cambridge university press, 2000 publication-title: Econom. Theory – volume: 33 start-page: 2734 year: 2020 end-page: 2746 ident: b45 article-title: Interventional few-shot learning publication-title: Adv. Neural Inf. Process. Syst. – reference: Lin, B., Zhang, S., Bao, F., 2020. Gait recognition with multiple-temporal-scale 3d convolutional neural network. In: Proceedings of the 28th ACM International Conference on Multimedia. pp. 3054–3062. – reference: Hu, Y., Wu, Y., Zhang, L., Wu, X., 2021. A generative adversarial framework for bounding confounded causal effects. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35, pp. 12104–12112. – volume: 150 start-page: 289 year: 2021 end-page: 296 ident: b42 article-title: Robust gait recognition using hybrid descriptors based on skeleton gait energy image publication-title: Pattern Recognit. Lett. – volume: Vol. 4 start-page: 441 year: 2006 end-page: 444 ident: b44 article-title: A framework for evaluating the effect of view angle, clothing and carrying condition on gait recognition publication-title: 18th International Conference on Pattern Recognition – reference: Zhang, Z., Tran, L., Yin, X., Atoum, Y., Liu, X., Wan, J., Wang, N., 2019b. Gait recognition via disentangled representation learning. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 4710–4719. – volume: 14 start-page: 102 year: 2018 end-page: 113 ident: b12 article-title: Multi-task GANs for view-specific feature learning in gait recognition publication-title: IEEE Trans. Inf. Forensics Secur. – reference: Chen, Z., Tian, Z., Zhu, J., Li, C., Du, S., 2022b. C-CAM: Causal CAM for Weakly Supervised Semantic Segmentation on Medical Image. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 11676–11685. – volume: 9 year: 2008 ident: b26 article-title: Visualizing data using t-SNE publication-title: J. Mach. Learn. Res. – reference: Staliūnaitė, I., Gorinski, P.J., Iacobacci, I., 2021. Improving commonsense causal reasoning by adversarial training and data augmentation. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 35, pp. 13834–13842. – reference: Li, B., Han, Z., Li, H., Fu, H., Zhang, C., 2022a. Trustworthy long-tailed classification. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 6970–6979. – year: 2017 ident: b13 article-title: In defense of the triplet loss for person re-identification – year: 2000 ident: b28 publication-title: Models, Reasoning and Inference – volume: 33 start-page: 3747 year: 2023 end-page: 3758 ident: b7 article-title: Counterfactual generation framework for few-shot learning publication-title: IEEE Trans. Circuits Syst. Video Technol. – volume: 10 start-page: 1 year: 2018 end-page: 14 ident: b31 article-title: Multi-view large population gait dataset and its performance evaluation for cross-view gait recognition publication-title: IPSJ Trans. Comput. Vis. Appl. – volume: 45 start-page: 12996 year: 2021 end-page: 13010 ident: b41 article-title: Deconfounded image captioning: A causal retrospect publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – reference: Zhu, B., Niu, Y., Hua, X.-S., Zhang, H., 2022. Cross-domain empirical risk minimization for unbiased long-tailed classification. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 36, pp. 3589–3597. – volume: 124 year: 2022 ident: b21 article-title: GaitSlice: A gait recognition model based on spatio-temporal slice features publication-title: Pattern Recognit. – volume: 14 start-page: 149 year: 2004 end-page: 158 ident: b35 article-title: Fusion of static and dynamic body biometrics for gait recognition publication-title: IEEE Trans. Circuits Syst. Video Technol. – volume: 39 start-page: 209 year: 2016 end-page: 226 ident: b37 article-title: A comprehensive study on cross-view gait based human identification with deep cnns publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – reference: Liu, B., Wang, D., Yang, X., Zhou, Y., Yao, R., Shao, Z., Zhao, J., 2022. Show, deconfound and tell: Image captioning with causal inference. In: Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. pp. 18041–18050. – volume: 31 start-page: 260 year: 2020 end-page: 274 ident: b39 article-title: Cross-view gait recognition using pairwise spatial transformer networks publication-title: IEEE Trans. Circuits Syst. Video Technol. – ident: 10.1016/j.cviu.2024.103995_b8 doi: 10.1109/CVPR52729.2023.00936 – ident: 10.1016/j.cviu.2024.103995_b24 doi: 10.1109/ICCV48922.2021.01438 – volume: 124 year: 2022 ident: 10.1016/j.cviu.2024.103995_b21 article-title: GaitSlice: A gait recognition model based on spatio-temporal slice features publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2021.108453 – ident: 10.1016/j.cviu.2024.103995_b4 doi: 10.1109/CVPR52688.2022.01138 – ident: 10.1016/j.cviu.2024.103995_b16 doi: 10.1609/aaai.v36i1.19983 – volume: 34 start-page: 21164 year: 2021 ident: 10.1016/j.cviu.2024.103995_b36 article-title: Provably efficient causal reinforcement learning with confounded observational data publication-title: Adv. Neural Inf. Process. Syst. – ident: 10.1016/j.cviu.2024.103995_b49 doi: 10.1109/CVPR46437.2021.00898 – volume: 9 issue: 11 year: 2008 ident: 10.1016/j.cviu.2024.103995_b26 article-title: Visualizing data using t-SNE publication-title: J. Mach. Learn. Res. – ident: 10.1016/j.cviu.2024.103995_b18 doi: 10.1109/CVPR52688.2022.00684 – volume: 34 start-page: 22905 year: 2021 ident: 10.1016/j.cviu.2024.103995_b29 article-title: Causal influence detection for improving efficiency in reinforcement learning publication-title: Adv. Neural Inf. Process. Syst. – ident: 10.1016/j.cviu.2024.103995_b6 doi: 10.1145/3503161.3547936 – volume: 636 year: 2023 ident: 10.1016/j.cviu.2024.103995_b5 article-title: GaitAMR: Cross-view gait recognition via aggregated multi-feature representation publication-title: Inform. Sci. doi: 10.1016/j.ins.2023.03.145 – volume: 45 start-page: 7639 issue: 6 year: 2022 ident: 10.1016/j.cviu.2024.103995_b38 article-title: PatchMix augmentation to identify causal features in few-shot learning publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2022.3223784 – volume: 14 start-page: 149 issue: 2 year: 2004 ident: 10.1016/j.cviu.2024.103995_b35 article-title: Fusion of static and dynamic body biometrics for gait recognition publication-title: IEEE Trans. Circuits Syst. Video Technol. doi: 10.1109/TCSVT.2003.821972 – volume: 150 start-page: 289 year: 2021 ident: 10.1016/j.cviu.2024.103995_b42 article-title: Robust gait recognition using hybrid descriptors based on skeleton gait energy image publication-title: Pattern Recognit. Lett. doi: 10.1016/j.patrec.2019.05.012 – start-page: 709 year: 2022 ident: 10.1016/j.cviu.2024.103995_b32 article-title: Invariant feature learning for generalized long-tailed classification – volume: 10 start-page: 1 year: 2018 ident: 10.1016/j.cviu.2024.103995_b31 article-title: Multi-view large population gait dataset and its performance evaluation for cross-view gait recognition publication-title: IPSJ Trans. Comput. Vis. Appl. – ident: 10.1016/j.cviu.2024.103995_b20 doi: 10.1109/CVPR42600.2020.01332 – ident: 10.1016/j.cviu.2024.103995_b33 doi: 10.1109/CVPRW56347.2022.00163 – ident: 10.1016/j.cviu.2024.103995_b2 doi: 10.1609/aaai.v33i01.33018126 – volume: 30 start-page: 3041 issue: 01 year: 2021 ident: 10.1016/j.cviu.2024.103995_b3 article-title: Multi-view gait image generation for cross-view gait recognition publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2021.3055936 – year: 2016 ident: 10.1016/j.cviu.2024.103995_b10 – volume: 45 start-page: 3933 issue: 3 year: 2022 ident: 10.1016/j.cviu.2024.103995_b34 article-title: Weakly-supervised video object grounding via causal intervention publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 28 start-page: 316 issue: 2 year: 2005 ident: 10.1016/j.cviu.2024.103995_b11 article-title: Individual recognition using gait energy image publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2006.38 – ident: 10.1016/j.cviu.2024.103995_b43 doi: 10.1109/CVPRW.2017.80 – ident: 10.1016/j.cviu.2024.103995_b25 doi: 10.1109/CVPR52688.2022.01751 – volume: 44 start-page: 345 issue: 1 year: 2020 ident: 10.1016/j.cviu.2024.103995_b47 article-title: On learning disentangled representations for gait recognition publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2020.2998790 – ident: 10.1016/j.cviu.2024.103995_b30 doi: 10.1609/aaai.v35i15.17630 – volume: 98 year: 2020 ident: 10.1016/j.cviu.2024.103995_b22 article-title: A model-based gait recognition method with body pose and human prior knowledge publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2019.107069 – volume: Vol. 4 start-page: 441 year: 2006 ident: 10.1016/j.cviu.2024.103995_b44 article-title: A framework for evaluating the effect of view angle, clothing and carrying condition on gait recognition – volume: 14 start-page: 102 issue: 1 year: 2018 ident: 10.1016/j.cviu.2024.103995_b12 article-title: Multi-task GANs for view-specific feature learning in gait recognition publication-title: IEEE Trans. Inf. Forensics Secur. doi: 10.1109/TIFS.2018.2844819 – ident: 10.1016/j.cviu.2024.103995_b1 doi: 10.1109/CVPR52688.2022.01961 – ident: 10.1016/j.cviu.2024.103995_b14 doi: 10.1609/aaai.v35i13.17437 – ident: 10.1016/j.cviu.2024.103995_b23 doi: 10.1145/3394171.3413861 – volume: 33 start-page: 2734 year: 2020 ident: 10.1016/j.cviu.2024.103995_b45 article-title: Interventional few-shot learning publication-title: Adv. Neural Inf. Process. Syst. – ident: 10.1016/j.cviu.2024.103995_b46 doi: 10.1109/CVPR.2019.00483 – volume: 33 start-page: 655 year: 2020 ident: 10.1016/j.cviu.2024.103995_b51 article-title: Causal intervention for weakly-supervised semantic segmentation publication-title: Adv. Neural Inf. Process. Syst. – volume: 45 start-page: 12996 issue: 11 year: 2021 ident: 10.1016/j.cviu.2024.103995_b41 article-title: Deconfounded image captioning: A causal retrospect publication-title: IEEE Trans. Pattern Anal. Mach. Intell. – volume: 32 start-page: 509 year: 2022 ident: 10.1016/j.cviu.2024.103995_b50 article-title: Learning domain invariant representations for generalizable person re-identification publication-title: IEEE Trans. Image Process. doi: 10.1109/TIP.2022.3229621 – ident: 10.1016/j.cviu.2024.103995_b9 doi: 10.1109/CVPR42600.2020.01423 – year: 2017 ident: 10.1016/j.cviu.2024.103995_b13 – volume: 37 start-page: 1057 issue: 5 year: 2004 ident: 10.1016/j.cviu.2024.103995_b40 article-title: Automated person recognition by walking and running via model-based approaches publication-title: Pattern Recognit. doi: 10.1016/j.patcog.2003.09.012 – volume: 33 start-page: 3747 issue: 8 year: 2023 ident: 10.1016/j.cviu.2024.103995_b7 article-title: Counterfactual generation framework for few-shot learning publication-title: IEEE Trans. Circuits Syst. Video Technol. doi: 10.1109/TCSVT.2023.3241651 – year: 2000 ident: 10.1016/j.cviu.2024.103995_b28 – volume: 32 start-page: 6967 issue: 10 year: 2022 ident: 10.1016/j.cviu.2024.103995_b15 article-title: Enhanced spatial-temporal salience for cross-view gait recognition publication-title: IEEE Trans. Circuits Syst. Video Technol. doi: 10.1109/TCSVT.2022.3175959 – ident: 10.1016/j.cviu.2024.103995_b17 doi: 10.1109/ICCV48922.2021.01267 – volume: 31 start-page: 260 issue: 1 year: 2020 ident: 10.1016/j.cviu.2024.103995_b39 article-title: Cross-view gait recognition using pairwise spatial transformer networks publication-title: IEEE Trans. Circuits Syst. Video Technol. doi: 10.1109/TCSVT.2020.2975671 – ident: 10.1016/j.cviu.2024.103995_b48 doi: 10.1109/CVPR.2019.00484 – ident: 10.1016/j.cviu.2024.103995_b19 doi: 10.1109/ICCVW54120.2021.00456 – volume: 39 start-page: 209 issue: 2 year: 2016 ident: 10.1016/j.cviu.2024.103995_b37 article-title: A comprehensive study on cross-view gait based human identification with deep cnns publication-title: IEEE Trans. Pattern Anal. Mach. Intell. doi: 10.1109/TPAMI.2016.2545669 – volume: 19 start-page: 675 issue: 4 year: 2003 ident: 10.1016/j.cviu.2024.103995_b27 article-title: Causality: models, reasoning, and inference, by judea pearl, cambridge university press, 2000 publication-title: Econom. Theory doi: 10.1017/S0266466603004109 – ident: 10.1016/j.cviu.2024.103995_b52 doi: 10.1609/aaai.v36i3.20271 |
| SSID | ssj0011491 |
| Score | 2.4432364 |
| Snippet | Gait recognition is a promising biometric technology that aims to identify the target subject via walking pattern. Most existing appearance-based methods focus... |
| SourceID | crossref elsevier |
| SourceType | Enrichment Source Index Database Publisher |
| StartPage | 103995 |
| SubjectTerms | Causal representation learning Disentangled representation learning Gait recognition Global and local feature extractor |
| Title | GaitSCM: Causal representation learning for gait recognition |
| URI | https://dx.doi.org/10.1016/j.cviu.2024.103995 |
| Volume | 243 |
| WOSCitedRecordID | wos001223459200001&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D |
| hasFullText | 1 |
| inHoldings | 1 |
| isFullTextHit | |
| isPrint | |
| journalDatabaseRights | – providerCode: PRVESC databaseName: Elsevier SD Freedom Collection Journals 2021 customDbUrl: eissn: 1090-235X dateEnd: 99991231 omitProxy: false ssIdentifier: ssj0011491 issn: 1077-3142 databaseCode: AIEXJ dateStart: 19950101 isFulltext: true titleUrlDefault: https://www.sciencedirect.com providerName: Elsevier |
| link | http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtV3Nb9MwFLfGxoEdYAzQNj7kA7cqlZPYiT1xmarBmESFRBG9RbFjo06QVWsz7c_fc20naYFpO3CJKst5jfr75fW95_eB0HvGKyVMrCIuwdOhJcsiSWMTqbyiKqkEU0Sthk3k4zGfTsVXn9a8WI0TyOua39yI-X-FGtYAbFs6-wC4W6GwAJ8BdLgC7HC9F_Cfytny2-iLdfVHZbNYNe6fd0VGdRgU4RIof8LuQZtF5DEKnQv8xIeBK0B35wy_bZJP0y-J6ZixCrv-0LMuSu81ScueiV85b-rNTeNAVB-DSGiXK-XVJslttJOu6dWEpj3NaI-c3TjNP5S2ix9cDNX1rBla8cNu83qH7I1_rjafMKSqXRRWRmFlFE7GI7ST5EyAvts5-Xw6PW9PmMAzjF0-qntyX1Dlcv82n-TvRkvPEJnsoafeg8AnDvnnaEvX--iZ9yaw19ULWArwhbV9tNvrPvkCffBMOcaOJ3idJzjwBANPsOUJ7vHkJfr-8XQyOov8MI1IpYQs4Q00JY9thCsVTGeaGCNImWdVboQ1-3JDE0mJrDgjJuU8U5KpSpcJWFHCZCR9hbbry1ofIJwYVlYM3mYtKWWSc8ErqSXJEiXAmuaHKA6_VqF8p3k78ORX8W-cDtGgvWfu-qzcuZsFEApvKToLsABO3XHf0YO-5TV60pH9DdpeXjX6LXqsrpezxdU7T6hbHA2Iqw |
| linkProvider | Elsevier |
| openUrl | ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=GaitSCM%3A+Causal+representation+learning+for+gait+recognition&rft.jtitle=Computer+vision+and+image+understanding&rft.au=Huo%2C+Wei&rft.au=Wang%2C+Ke&rft.au=Tang%2C+Jun&rft.au=Wang%2C+Nian&rft.date=2024-06-01&rft.issn=1077-3142&rft.volume=243&rft.spage=103995&rft_id=info:doi/10.1016%2Fj.cviu.2024.103995&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_cviu_2024_103995 |
| thumbnail_l | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=1077-3142&client=summon |
| thumbnail_m | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=1077-3142&client=summon |
| thumbnail_s | http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=1077-3142&client=summon |