You Only Erase Once: Erasing Anything without Bringing Unexpected Content
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866908919470751744 |
|---|---|
| author | Zhu, Yixing Zhang, Qing Xu, Wenju Zheng, Wei-Shi |
| author_facet | Zhu, Yixing Zhang, Qing Xu, Wenju Zheng, Wei-Shi |
| contents | We present YOEO, an approach for object erasure. Unlike recent diffusion-based methods which struggle to erase target objects without generating unexpected content within the masked regions due to lack of sufficient paired training data and explicit constraint on content generation, our method allows to produce high-quality object erasure results free of unwanted objects or artifacts while faithfully preserving the overall context coherence to the surrounding content. We achieve this goal by training an object erasure diffusion model on unpaired data containing only large-scale real-world images, under the supervision of a sundries detector and a context coherence loss that are built upon an entity segmentation model. To enable more efficient training and inference, a diffusion distillation strategy is employed to train for a few-step erasure diffusion model. Extensive experiments show that our method outperforms the state-of-the-art object erasure methods. Code will be available at https://zyxunh.github.io/YOEO-ProjectPage/. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_27599 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | You Only Erase Once: Erasing Anything without Bringing Unexpected Content Zhu, Yixing Zhang, Qing Xu, Wenju Zheng, Wei-Shi Computer Vision and Pattern Recognition We present YOEO, an approach for object erasure. Unlike recent diffusion-based methods which struggle to erase target objects without generating unexpected content within the masked regions due to lack of sufficient paired training data and explicit constraint on content generation, our method allows to produce high-quality object erasure results free of unwanted objects or artifacts while faithfully preserving the overall context coherence to the surrounding content. We achieve this goal by training an object erasure diffusion model on unpaired data containing only large-scale real-world images, under the supervision of a sundries detector and a context coherence loss that are built upon an entity segmentation model. To enable more efficient training and inference, a diffusion distillation strategy is employed to train for a few-step erasure diffusion model. Extensive experiments show that our method outperforms the state-of-the-art object erasure methods. Code will be available at https://zyxunh.github.io/YOEO-ProjectPage/. |
| title | You Only Erase Once: Erasing Anything without Bringing Unexpected Content |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2603.27599 |