Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2605.09003 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866917483061968896 |
|---|---|
| author | Tang, Yixin Guo, Jiawei Li, Junxian Li, Zhiteng Zhao, Jixin Zhang, Bingya Wang, Chenbo Zhang, Yulun Zhou, Shangchen |
| author_facet | Tang, Yixin Guo, Jiawei Li, Junxian Li, Zhiteng Zhao, Jixin Zhang, Bingya Wang, Chenbo Zhang, Yulun Zhou, Shangchen |
| contents | Recently, diffusion-based object removal models have achieved impressive results in eliminating objects and their associated visual effects. However, they indiscriminately denoise all tokens across all timesteps, ignoring that removal usually involves small foreground regions. This strategy introduces substantial computational overhead and prolonged inference times. To overcome this computational burden, we propose a latent discriminator to implement Region-aware Adversarial Distillation (RAD), yielding a highly efficient few-step model named FlashClear. Furthermore, tailored to few-step diffusion models, we propose FPAC (Foreground-Prioritized Asymmetric Attention and Caching), a training-free acceleration strategy. Extensive experiments demonstrate that our framework provides massive acceleration while maintaining or exceeding the performance of our base model, ObjectClear. Notably, on the OBER benchmark, our FlashClear achieves up to 8.26$\times$ and 122$\times$ speedup over ObjectClear and OmniPaint, respectively, while maintaining high visual quality and fidelity. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2605_09003 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | FlashClear: Ultra-Fast Image Content Removal via Efficient Step Distillation and Feature Caching Tang, Yixin Guo, Jiawei Li, Junxian Li, Zhiteng Zhao, Jixin Zhang, Bingya Wang, Chenbo Zhang, Yulun Zhou, Shangchen Computer Vision and Pattern Recognition Recently, diffusion-based object removal models have achieved impressive results in eliminating objects and their associated visual effects. However, they indiscriminately denoise all tokens across all timesteps, ignoring that removal usually involves small foreground regions. This strategy introduces substantial computational overhead and prolonged inference times. To overcome this computational burden, we propose a latent discriminator to implement Region-aware Adversarial Distillation (RAD), yielding a highly efficient few-step model named FlashClear. Furthermore, tailored to few-step diffusion models, we propose FPAC (Foreground-Prioritized Asymmetric Attention and Caching), a training-free acceleration strategy. Extensive experiments demonstrate that our framework provides massive acceleration while maintaining or exceeding the performance of our base model, ObjectClear. Notably, on the OBER benchmark, our FlashClear achieves up to 8.26$\times$ and 122$\times$ speedup over ObjectClear and OmniPaint, respectively, while maintaining high visual quality and fidelity. |
| title | FlashClear: Ultra-Fast Image Content Removal via Efficient Step Distillation and Feature Caching |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2605.09003 |