Few Shot Retail Shelf Segmentation with Coarse Prototype Matching Episodes
DOI:
https://doi.org/10.66238/jaeee25Keywords:
Retail Automation, Few-Shot Segmentation, Prototype Matching, Computer VisionAbstract
The rapid evolution of automated retail systems has necessitated the development of highly accurate and adaptable computer vision algorithms capable of parsing complex shelf environments. Retail shelf segmentation remains an exceptionally challenging domain due to extreme product density, severe intra-class variations, inter-class similarities, and dynamic illumination conditions. Furthermore, the constant introduction of new stock keeping units makes traditional fully supervised learning paradigms impractical, as they require exhaustive pixel-level annotations for every new product. To address these systemic bottlenecks, this paper proposes a novel coarse-to-fine prototype matching framework specifically tailored for episodic tasks in few-shot retail shelf segmentation. By leveraging the episodic training paradigm, the proposed architecture learns to generalize to unseen product categories using only a minimal number of annotated support samples. Our methodology introduces a dual-phase prototype extraction mechanism. Initially, a coarse global prototype establishes a baseline representation of the target category, facilitating broad localization and mitigating background noise. Subsequently, a fine-grained local prototype matching module captures intricate structural and textural variations across different spatial regions of the products. This hierarchical approach effectively bridges the semantic gap between the support and query sets in highly cluttered scenes. Extensive evaluations on a newly curated retail shelf segmentation dataset demonstrate that the proposed coarse-to-fine framework significantly outperforms existing state-of-the-art methods in both one-shot and five-shot scenarios. The findings indicate substantial improvements in boundary delineations and robustness against heavy occlusions, marking a critical advancement toward scalable and autonomous inventory management systems.
References
1. Guo, Hanzhong, et al. "Leveraging verifier-based reinforcement learning in image editing." arXiv preprint arXiv:2604.27505 (2026).
2. Li, Q., Luo, T., & Liao, J. (2025). Theory-inspired deep multi-view multi-label learning with incomplete views and noisy labels. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 20706-20715).
3. Tang, A., Hiraoka, T., Hiraoka, N., Shi, F., Kawaharazuka, K., Kojima, K., ... & Inaba, M. (2024, May). Humanmimic: Learning natural locomotion and transitions for humanoid robot via wasserstein adversarial imitation. In 2024 IEEE International Conference on Robotics and Automation (ICRA) (pp. 13107-13114). IEEE.
4. Lv, Q., Kong, W., Li, H., Zeng, J., Qiu, Z., Qu, D., ... & Pang, J. (2025). F1: A vision-language-action model bridging understanding and generation to actions. arXiv preprint arXiv:2509.06951.
5. Yang, S., Liu, F., Chen, D., Huang, H., & Zhou, J. (2024). Few-shot classification model compression via school learning. IEEE Transactions on Circuits and Systems for Video Technology, 34(12), 12244-12257.
6. Sang, Y. (2025, August). Robustness of fine-tuned llms under noisy retrieval inputs. In 2025 6th International Conference on Artificial Intelligence and Electromechanical Automation (AIEA) (pp. 417-420). IEEE.
7. Zhao, C., Zhang, J., Du, J., Shan, Z., Wang, J., Yu, J., ... & Xu, L. (2024). I'm hoi: Inertia-aware monocular capture of 3d human-object interactions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 729-741).
8. Guo, C., Xiao, Z., Liu, Z., & Peng, S. (2020). Rg conditional diagnosability: a novel generalized measure of system-level diagnosis. Theoretical Computer Science, 814, 19-27.
9. Zhang, J., Luo, H., Yang, H., Xu, X., Wu, Q., Shi, Y., ... & Wang, J. (2023). Neuraldome: A neural modeling pipeline on multi-view human-object interactions. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 8834-8845).
10. Guo, H., Lu, C., Bao, F., Pang, T., Yan, S., Du, C., & Li, C. (2023). Gaussian mixture solvers for diffusion models. Advances in Neural Information Processing Systems, 36, 25598-25626.
11. Yang, Y., Ma, X., Li, C., Zheng, Z., Zhang, Q., Huang, G., ... & Zhao, Q. (2021). Believe what you see: Implicit constraint approach for offline multi-agent reinforcement learning. Advances in Neural Information Processing Systems, 34, 10299-10312.
12. Liu, Y., Liu, M., Huang, S., & Lv, J. (2025, April). Asymmetric visual semantic embedding framework for efficient vision-language alignment. In Proceedings of the AAAI Conference on Artificial Intelligence (Vol. 39, No. 6, pp. 5676-5684).
13. Chu, Y. B., Li, J., Jia, P., Cui, J., Zhang, R., Kang, X., Lv, M., & Zhang, S. (2021). Irf1- and Egr1-activated transcription plays a key role in macrophage polarization: A multiomics sequencing study with partial validation. International immunopharmacology, 99, 108072. https://doi.org/10.1016/j.intimp.2021.108072
14. Peng, Q., Zheng, C., & Chen, C. (2024). A dual-augmentor framework for domain generalization in 3d human pose estimation. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 2240-2249).
15. Qi, Z., Wang, S., Zhang, W., & Huang, Q. (2024). Uncertainty-boosted robust video activity anticipation. IEEE Transactions on Pattern Analysis and Machine Intelligence, 46(12), 7775-7792.
16. He, Y., Li, S., Li, K., Wang, J., Li, B., Shi, T., ... & Wang, X. (2025). Ge-adapter: A general and efficient adapter for enhanced video editing with pretrained text-to-image diffusion models. Expert Systems with Applications, 129649.
17. Mi, L., Wang, W., Tu, W., He, Q., Kong, R., Fang, X., ... & Liu, Y. (2025, March). Empower vision applications with LoRA LMM. In Proceedings of the Twentieth European Conference on Computer Systems (pp. 261-277).
18. Zhao, H., Wang, Q., Zhan, G., Min, W., Zou, Y., & Cui, S. (2022). Need only one more point (NOOMP): Perspective adaptation crowd counting in complex scenes. IEEE Transactions on Multimedia, 25, 1414-1426.
19. Mi, L., Yuan, T., Wang, W., Dai, H., Sun, L., Zheng, J., ... & Fu, X. (2024). Accelerated neural enhancement for video analytics with video quality adaptation. IEEE/ACM Transactions on Networking, 32(4), 3045-3060.
20. Li, Q., Luo, T., Jiang, M., Liao, J., & Jiang, Z. (2024, October). Deep incomplete multi-view network semi-supervised multi-label learning with unbiased loss. In Proceedings of the 32nd ACM International Conference on Multimedia (pp. 9048-9056).
21. Zou, Fanglei, et al. "Enhancing cyclodextrin glycosyltransferase-mediated potato starch modification via plasma-activated water and twin-screw extrusion treatment." Food Chemistry 491 (2025): 145247.
22. Huang, Z., Wang, J., Chen, L., Xiao, B., Cai, L., Zeng, Y., & Xu, J. (2025, October). MVISU-Bench: Benchmarking Mobile Agents for Real-World Tasks by Multi-App, Vague, Interactive, Single-App and Unethical Instructions. In Proceedings of the 33rd ACM International Conference on Multimedia (pp. 8797-8805).
23. Ma, Y., Qu, D., & Pyrozhenko, M. (2026). Bio-RegNet: A Meta-Homeostatic Bayesian Neural Network Framework Integrating Treg-Inspired Immunoregulation and Autophagic Optimization for Adaptive Community Detection and Stable Intelligence. Biomimetics, 11(1), 48.
24. Qu, W., Wang, J., Gong, Y., Huang, X., & Xiao, L. (2025). An end-to-end robust point cloud semantic segmentation network with single-step conditional diffusion models. In Proceedings of the Computer Vision and Pattern Recognition Conference (pp. 27325-27335).
25. Jiang, L., Cao, Z., Chen, W., Zheng, X., Lyu, Y., Li, Z., ... & Chen, Y. C. (2026). SAP: Segment Any 4K Panorama. arXiv preprint arXiv:2603.12759.
26. Zhang, B., Wang, X., Kao, H., Zhang, G., Gao, C., Wang, Y., ... & Chen, K. (2026). Consensus Matrix: A Role-Specialized Multi-Agent Framework for Structured Collaborative Decision-Making in Agentic Visual Media Workflows. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 4613-4622).
27. Tu, P., Huang, Y., Zheng, F., He, Z., Cao, L., & Shao, L. (2022, June). Guidedmix-net: Semi-supervised semantic segmentation by using labeled images as reference. In Proceedings of the AAAI conference on artificial intelligence (Vol. 36, No. 2, pp. 2379-2387).
28. Qi, Z., Yuan, Y., Ruan, X., Wang, S., Zhang, W., & Huang, Q. (2024). Collaborative debias strategy for temporal sentence grounding in video. IEEE Transactions on Circuits and Systems for Video Technology, 34(11), 10972-10986.
29. Huang, H., Zhang, J., Zhang, J., Xu, J., & Wu, Q. (2020). Low-rank pairwise alignment bilinear network for few-shot fine-grained image classification. IEEE Transactions on Multimedia, 23, 1666-1680.
30. Li, Q., Liu, Z., Xu, T., Luo, T., & Hou, C. (2026). Adaptive Disentangled Representation Learning for Incomplete Multi-View Multi-Label Classification. arXiv preprint arXiv:2601.05785.
31. Qu, W., Shao, Y., Meng, L., Huang, X., & Xiao, L. (2024). A conditional denoising diffusion probabilistic model for point cloud upsampling. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 20786-20795).
32. Huang, H., Kang, H., Liu, S., Salvado, O., Rakotoarivelo, T., Wang, D., & Liu, T. (2023). Paddles: Phase-amplitude spectrum disentangled early stopping for learning with noisy labels. In Proceedings of the IEEE/CVF International Conference on Computer Vision (pp. 16719-16730).
33. Xu, X., Bowtell, J., Young, W. R., Fong, D. T., & Williams, G. K. (2025). Distinct patterns in neuromuscular adaptation to repeated perturbations in chronic ankle instability. Journal of NeuroEngineering and Rehabilitation.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Joonseo Hwang
This work is licensed under a Creative Commons Attribution 4.0 International License.
All articles published in the Journal of Advanced Electrical and Electronic Engineering (JAEEE) are distributed under the Creative Commons Attribution 4.0 International License (CC BY 4.0). Authors retain copyright of their work while granting the journal the right of first publication.
Under this license, users are permitted to read, download, copy, distribute, print, share, adapt, and reuse published material for both commercial and non-commercial purposes, provided appropriate credit is given to the original author(s), the source of publication is acknowledged, a link to the license is provided, and any changes made are clearly indicated.
The license does not permit users to imply endorsement by the original authors or the journal. No additional restrictions may be imposed that prevent others from exercising the rights granted under CC BY 4.0.
By submitting to the journal, authors confirm that their work is original, does not infringe upon third-party rights, and complies with ethical publishing standards.
