Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866929706606002176 |
|---|---|
| author | Hu, Li Wang, Guangyuan Shen, Zhen Gao, Xin Meng, Dechao Zhuo, Lian Zhang, Peng Zhang, Bang Bo, Liefeng |
| author_facet | Hu, Li Wang, Guangyuan Shen, Zhen Gao, Xin Meng, Dechao Zhuo, Lian Zhang, Peng Zhang, Bang Bo, Liefeng |
| contents | Recent character image animation methods based on diffusion models, such as Animate Anyone, have made significant progress in generating consistent and generalizable character animations. However, these approaches fail to produce reasonable associations between characters and their environments. To address this limitation, we introduce Animate Anyone 2, aiming to animate characters with environment affordance. Beyond extracting motion signals from source video, we additionally capture environmental representations as conditional inputs. The environment is formulated as the region with the exclusion of characters and our model generates characters to populate these regions while maintaining coherence with the environmental context. We propose a shape-agnostic mask strategy that more effectively characterizes the relationship between character and environment. Furthermore, to enhance the fidelity of object interactions, we leverage an object guider to extract features of interacting objects and employ spatial blending for feature injection. We also introduce a pose modulation strategy that enables the model to handle more diverse motion patterns. Experimental results demonstrate the superior performance of the proposed method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_06145 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance Hu, Li Wang, Guangyuan Shen, Zhen Gao, Xin Meng, Dechao Zhuo, Lian Zhang, Peng Zhang, Bang Bo, Liefeng Computer Vision and Pattern Recognition Recent character image animation methods based on diffusion models, such as Animate Anyone, have made significant progress in generating consistent and generalizable character animations. However, these approaches fail to produce reasonable associations between characters and their environments. To address this limitation, we introduce Animate Anyone 2, aiming to animate characters with environment affordance. Beyond extracting motion signals from source video, we additionally capture environmental representations as conditional inputs. The environment is formulated as the region with the exclusion of characters and our model generates characters to populate these regions while maintaining coherence with the environmental context. We propose a shape-agnostic mask strategy that more effectively characterizes the relationship between character and environment. Furthermore, to enhance the fidelity of object interactions, we leverage an object guider to extract features of interacting objects and employ spatial blending for feature injection. We also introduce a pose modulation strategy that enables the model to handle more diverse motion patterns. Experimental results demonstrate the superior performance of the proposed method. |
| title | Animate Anyone 2: High-Fidelity Character Image Animation with Environment Affordance |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2502.06145 |