Saved in:
| Main Authors: | , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2510.07721 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916998027411456 |
|---|---|
| author | Guo, Zipeng Ma, Lichen Fu, Xiaolong Zhou, Gaojing Yang, Lan Zhou, Yuchen Liu, Linkai He, Yu Liu, Ximan Dong, Shiping Fu, Jingling Chen, Zhen Shi, Yu Huang, Junshi Li, Jason Gou, Chao |
| author_facet | Guo, Zipeng Ma, Lichen Fu, Xiaolong Zhou, Gaojing Yang, Lan Zhou, Yuchen Liu, Linkai He, Yu Liu, Ximan Dong, Shiping Fu, Jingling Chen, Zhen Shi, Yu Huang, Junshi Li, Jason Gou, Chao |
| contents | In web data, product images are central to boosting user engagement and advertising efficacy on e-commerce platforms, yet the intrusive elements such as watermarks and promotional text remain major obstacles to delivering clear and appealing product visuals. Although diffusion-based inpainting methods have advanced, they still face challenges in commercial settings due to unreliable object removal and limited domain-specific adaptation. To tackle these challenges, we propose Repainter, a reinforcement learning framework that integrates spatial-matting trajectory refinement with Group Relative Policy Optimization (GRPO). Our approach modulates attention mechanisms to emphasize background context, generating higher-reward samples and reducing unwanted object insertion. We also introduce a composite reward mechanism that balances global, local, and semantic constraints, effectively reducing visual artifacts and reward hacking. Additionally, we contribute EcomPaint-100K, a high-quality, large-scale e-commerce inpainting dataset, and a standardized benchmark EcomPaint-Bench for fair evaluation. Extensive experiments demonstrate that Repainter significantly outperforms state-of-the-art methods, especially in challenging scenes with intricate compositions. We will release our code and weights upon acceptance. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_07721 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | RePainter: Empowering E-commerce Object Removal via Spatial-matting Reinforcement Learning Guo, Zipeng Ma, Lichen Fu, Xiaolong Zhou, Gaojing Yang, Lan Zhou, Yuchen Liu, Linkai He, Yu Liu, Ximan Dong, Shiping Fu, Jingling Chen, Zhen Shi, Yu Huang, Junshi Li, Jason Gou, Chao Computer Vision and Pattern Recognition In web data, product images are central to boosting user engagement and advertising efficacy on e-commerce platforms, yet the intrusive elements such as watermarks and promotional text remain major obstacles to delivering clear and appealing product visuals. Although diffusion-based inpainting methods have advanced, they still face challenges in commercial settings due to unreliable object removal and limited domain-specific adaptation. To tackle these challenges, we propose Repainter, a reinforcement learning framework that integrates spatial-matting trajectory refinement with Group Relative Policy Optimization (GRPO). Our approach modulates attention mechanisms to emphasize background context, generating higher-reward samples and reducing unwanted object insertion. We also introduce a composite reward mechanism that balances global, local, and semantic constraints, effectively reducing visual artifacts and reward hacking. Additionally, we contribute EcomPaint-100K, a high-quality, large-scale e-commerce inpainting dataset, and a standardized benchmark EcomPaint-Bench for fair evaluation. Extensive experiments demonstrate that Repainter significantly outperforms state-of-the-art methods, especially in challenging scenes with intricate compositions. We will release our code and weights upon acceptance. |
| title | RePainter: Empowering E-commerce Object Removal via Spatial-matting Reinforcement Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.07721 |