BlobCtrl: Taming Controllable Blob for Element-level Image Editing
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909817164005376 |
|---|---|
| author | Li, Yaowei Li, Lingen Zhang, Zhaoyang Li, Xiaoyu Wang, Guangzhi Li, Hongxiang Cun, Xiaodong Shan, Ying Zou, Yuexian |
| author_facet | Li, Yaowei Li, Lingen Zhang, Zhaoyang Li, Xiaoyu Wang, Guangzhi Li, Hongxiang Cun, Xiaodong Shan, Ying Zou, Yuexian |
| contents | As user expectations for image editing continue to rise, the demand for flexible, fine-grained manipulation of specific visual elements presents a challenge for current diffusion-based methods. In this work, we present BlobCtrl, a framework for element-level image editing based on a probabilistic blob-based representation. Treating blobs as visual primitives, BlobCtrl disentangles layout from appearance, affording fine-grained, controllable object-level manipulation. Our key contributions are twofold: (1) an in-context dual-branch diffusion model that separates foreground and background processing, incorporating blob representations to explicitly decouple layout and appearance, and (2) a self-supervised disentangle-then-reconstruct training paradigm with an identity-preserving loss function, along with tailored strategies to efficiently leverage blob-image pairs. To foster further research, we introduce BlobData for large-scale training and BlobBench, a benchmark for systematic evaluation. Experimental results demonstrate that BlobCtrl achieves state-of-the-art performance in a variety of element-level editing tasks, such as object addition, removal, scaling, and replacement, while maintaining computational efficiency. Project Webpage: https://liyaowei-stu.github.io/project/BlobCtrl/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_13434 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | BlobCtrl: Taming Controllable Blob for Element-level Image Editing Li, Yaowei Li, Lingen Zhang, Zhaoyang Li, Xiaoyu Wang, Guangzhi Li, Hongxiang Cun, Xiaodong Shan, Ying Zou, Yuexian Computer Vision and Pattern Recognition Artificial Intelligence Multimedia As user expectations for image editing continue to rise, the demand for flexible, fine-grained manipulation of specific visual elements presents a challenge for current diffusion-based methods. In this work, we present BlobCtrl, a framework for element-level image editing based on a probabilistic blob-based representation. Treating blobs as visual primitives, BlobCtrl disentangles layout from appearance, affording fine-grained, controllable object-level manipulation. Our key contributions are twofold: (1) an in-context dual-branch diffusion model that separates foreground and background processing, incorporating blob representations to explicitly decouple layout and appearance, and (2) a self-supervised disentangle-then-reconstruct training paradigm with an identity-preserving loss function, along with tailored strategies to efficiently leverage blob-image pairs. To foster further research, we introduce BlobData for large-scale training and BlobBench, a benchmark for systematic evaluation. Experimental results demonstrate that BlobCtrl achieves state-of-the-art performance in a variety of element-level editing tasks, such as object addition, removal, scaling, and replacement, while maintaining computational efficiency. Project Webpage: https://liyaowei-stu.github.io/project/BlobCtrl/ |
| title | BlobCtrl: Taming Controllable Blob for Element-level Image Editing |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Multimedia |
| url | https://arxiv.org/abs/2503.13434 |