MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesis
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866929613533347840 |
|---|---|
| author | Wang, Tianyu Zhang, Jianming Zheng, Haitian Ding, Zhihong Cohen, Scott Lin, Zhe Xiong, Wei Fu, Chi-Wing Figueroa, Luis Kim, Soo Ye |
| author_facet | Wang, Tianyu Zhang, Jianming Zheng, Haitian Ding, Zhihong Cohen, Scott Lin, Zhe Xiong, Wei Fu, Chi-Wing Figueroa, Luis Kim, Soo Ye |
| contents | Shadows are often under-considered or even ignored in image editing applications, limiting the realism of the edited results. In this paper, we introduce MetaShadow, a three-in-one versatile framework that enables detection, removal, and controllable synthesis of shadows in natural images in an object-centered fashion. MetaShadow combines the strengths of two cooperative components: Shadow Analyzer, for object-centered shadow detection and removal, and Shadow Synthesizer, for reference-based controllable shadow synthesis. Notably, we optimize the learning of the intermediate features from Shadow Analyzer to guide Shadow Synthesizer to generate more realistic shadows that blend seamlessly with the scene. Extensive evaluations on multiple shadow benchmark datasets show significant improvements of MetaShadow over the existing state-of-the-art methods on object-centered shadow detection, removal, and synthesis. MetaShadow excels in image-editing tasks such as object removal, relocation, and insertion, pushing the boundaries of object-centered image editing. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_02635 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesis Wang, Tianyu Zhang, Jianming Zheng, Haitian Ding, Zhihong Cohen, Scott Lin, Zhe Xiong, Wei Fu, Chi-Wing Figueroa, Luis Kim, Soo Ye Computer Vision and Pattern Recognition Shadows are often under-considered or even ignored in image editing applications, limiting the realism of the edited results. In this paper, we introduce MetaShadow, a three-in-one versatile framework that enables detection, removal, and controllable synthesis of shadows in natural images in an object-centered fashion. MetaShadow combines the strengths of two cooperative components: Shadow Analyzer, for object-centered shadow detection and removal, and Shadow Synthesizer, for reference-based controllable shadow synthesis. Notably, we optimize the learning of the intermediate features from Shadow Analyzer to guide Shadow Synthesizer to generate more realistic shadows that blend seamlessly with the scene. Extensive evaluations on multiple shadow benchmark datasets show significant improvements of MetaShadow over the existing state-of-the-art methods on object-centered shadow detection, removal, and synthesis. MetaShadow excels in image-editing tasks such as object removal, relocation, and insertion, pushing the boundaries of object-centered image editing. |
| title | MetaShadow: Object-Centered Shadow Detection, Removal, and Synthesis |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.02635 |