Research on Fashion Contextual Style Recognition Methods Based on Scene Graph Technology
DOI:
https://doi.org/10.64509/jdi.13.115Keywords:
Scene Graph Generation, Fashion Contextual Style Recognition, Unbiased Scene Graph, Hierarchical Context, Personalised RecommendationsAbstract
Driven by the intelligent and personalized transformation of the garment industry, fashion contextual style recognition has emerged as a pivotal technology bridging visual understanding and business decision-making. However, traditional methods anchor styles to isolated garment attributes, failing to capture the deep semantics of entity interactions in complex visual scenes. To address these limitations, we propose a fashion contextual style recognition method based on unbiased Scene Graph Generation (SGG) and hierarchical context fusion. We developed a hierarchical context-aware network that deconstructs fashion contexts into three dimensions: entity context, global context, and scene context, integrated through graph embedding. Furthermore, a Causal Intervention mechanism based on Total Direct Effect (TDE) is introduced to mitigate the long-tail bias inherent in unconstrained scenes. Experimental results on the Visual Genome dataset demonstrate that our approach significantly outperforms baseline models in recognition accuracy and robustness. This research extends the definition of fashion style from visual attributes to holistic situational semantics, providing a new perspective for intelligent fashion analysis.
Downloads
References
[1] Sun, G.- L., Cheng, Z.- Q., Wu, X., Peng, Q.: Personalized clothing recommendation combining user social circle and fashion style consistency. Multimedia Tools and Applications 77, 17731-17754 (2018). https://doi.org/10.1007/s11042-017-5245-1
[2] Lee, H., Seol, J., Lee, S.- g.: Style2vec: Representation learning for fashion items from style sets. arXiv preprint arXiv:1708.04014 (2017). https://doi.org/10.48550/arXiv.1708.04014
[3] Yue, X., Zhang, C., Fujita, H., Lv, Y.: Clothing fashion style recognition with design issue graph. Applied Intelligence 51(6), 3548-3560 (2021). https://doi.org/10.1007/s10489-020-01950-7
[4] Bossard, L., Dantone, M., Leistner, C., Wengert, C., Quack, T., Van Gool, L.: Apparel classification with style. In Proceedings of the 11th Asian Conference on Computer Vision, pp. 321-335. https://doi.org/10.1007/978-3-642-37447-0_25
[5] Ge, Y., Zhang, R., Wang, X., Tang, X., Luo, P.: DeepFashion2: A Versatile Benchmark for Detection, Pose Estimation, Segmentation and Re-Identification of Clothing Images. In 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 5332-5340 (2019). https://doi.org/10.1109/CVPR.2019.00548
[6] Vaccaro, K., Shivakumar, S., Ding, Z., Karahalios, K., Kumar, R.: The elements of fashion style. In Proceedings of the 29th Annual Symposium on User Interface Software and Technology, pp. 777-785. https://doi.org/10.1145/2984511.2984573
[7] Kiapour, M.H., Yamaguchi, K., Berg, A.C., Berg, T.L.: Hipster wars: Discovering elements of fashion styles. In European Conference on Computer Vision, pp. 472-488. https://doi.org/10.1007/978-3-319-10590-1_31
[8] Krishna, R., Zhu, Y., Groth, O., Johnson, J., Hata, K., Kravitz, J., Chen, S., Kalantidis, Y., Li, L.- J., Shamma, D.A.: Visual genome: Connecting language and vision using crowdsourced dense image annotations. International journal of computer vision 123(1), 32-73 (2017). https://doi.org/10.1007/s11263-016-0981-7
[9] Yang, J., Lu, J., Lee, S., Batra, D., Parikh, D.: Graph RCNN for Scene Graph Generation. In Proceedings of the European Conference on Computer Vision (ECCV), pp. 670-685. https://doi.org/10.1007/978-3-030-01246-5_41
[10] Wu, S., Zhou, L., Hu, Z., Liu, J.: Hierarchical context-based emotion recognition with scene graphs. IEEE Transactions on Neural Networks and Learning Systems 35(3), 3725-3739 (2022). https://doi.org/10.1109/TNNLS.2022.3193837
[11] Tang, K., Niu, Y., Huang, J., Shi, J., Zhang, H.: Unbiased scene graph generation from biased training. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, pp. 3716-3725. https://doi.org/10.1109/CVPR42600.2020.00377
[12] Tolstikhin, I.O., Houlsby, N., Kolesnikov, A., Beyer, L., Zhai, X., Unterthiner, T., Yung, J., Steiner, A., Keysers, D., Uszkoreit, J.: Mlp-mixer: An all-mlp architecture for vision. In Proceedings of the 35th International Conference on Neural Information Processing Systems, pp. 24261-24272 (2021)
[13] Godart, F.C.: "Why is style not in fashion? Using the concept of "style" to understand the creative industries", in Frontiers of Creative Industries: Exploring Structural and Categorical Dynamics. Emerald Publishing Limited, Leeds, United Kingdom (2018). https://doi.org/10.1108/S0733-558X20180000055005
[14] Wang, K., Zhang, J., Zhang, P., Sun, K., Zhan, J., Wei, M.: Personal Style Guided Outfit Recommendation with Multi-Modal Fashion Compatibility Modeling. Journal of Donghua University(English Edition) 42(02), 156-167 (2025). https://doi.org/10.19884/j.1672-5220.202404017
[15] Dey, A.U., Ghosh, S.K., Valveny, E., Harit, G.: Beyond visual semantics: Exploring the role of scene text in image understanding. Pattern Recognition Letters 149, 164-171 (2021). https://doi.org/10.1016/j.patrec.2021.06.011
[16] Wu, Z., Pan, S., Chen, F., Long, G., Zhang, C., Yu, P.S.: A comprehensive survey on graph neural networks. IEEE transactions on neural networks and learning systems 32(1), 4-24 (2020). https://doi.org/10.1109/TNNLS.2020.2978386
[17] Finn, C., Abbeel, P., Levine, S.: Model-agnostic meta-learning for fast adaptation of deep networks. In Proceedings of the 34th International Conference on Machine Learning, pp. 1126-1135
[18] Velickovic, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., Bengio, Y.: Graph attention networks. arXiv preprint arXiv:1710.10903 (2017). https://doi.org/10.48550/arXiv.1710.10903
[19] Hinton, G., Vinyals, O., Dean, J.: Distilling the knowledge in a neural network. arXiv preprint arXiv:1503.02531 (2015). https://doi.org/10.48550/arXiv.1503.02531
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Authors
This work is licensed under a Creative Commons Attribution 4.0 International License.
