TweedieMix: Improving Multi-Concept Fusion for Diffusion-based Image/Video Generation
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_ | 1866916641725480960 |
|---|---|
| author | Kwon, Gihyun Ye, Jong Chul |
| author_facet | Kwon, Gihyun Ye, Jong Chul |
| contents | Despite significant advancements in customizing text-to-image and video generation models, generating images and videos that effectively integrate multiple personalized concepts remains a challenging task. To address this, we present TweedieMix, a novel method for composing customized diffusion models during the inference phase. By analyzing the properties of reverse diffusion sampling, our approach divides the sampling process into two stages. During the initial steps, we apply a multiple object-aware sampling technique to ensure the inclusion of the desired target objects. In the later steps, we blend the appearances of the custom concepts in the de-noised image space using Tweedie's formula. Our results demonstrate that TweedieMix can generate multiple personalized concepts with higher fidelity than existing methods. Moreover, our framework can be effortlessly extended to image-to-video diffusion models, enabling the generation of videos that feature multiple personalized concepts. Results and source code are in our anonymous project page. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_05591 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | TweedieMix: Improving Multi-Concept Fusion for Diffusion-based Image/Video Generation Kwon, Gihyun Ye, Jong Chul Computer Vision and Pattern Recognition Despite significant advancements in customizing text-to-image and video generation models, generating images and videos that effectively integrate multiple personalized concepts remains a challenging task. To address this, we present TweedieMix, a novel method for composing customized diffusion models during the inference phase. By analyzing the properties of reverse diffusion sampling, our approach divides the sampling process into two stages. During the initial steps, we apply a multiple object-aware sampling technique to ensure the inclusion of the desired target objects. In the later steps, we blend the appearances of the custom concepts in the de-noised image space using Tweedie's formula. Our results demonstrate that TweedieMix can generate multiple personalized concepts with higher fidelity than existing methods. Moreover, our framework can be effortlessly extended to image-to-video diffusion models, enabling the generation of videos that feature multiple personalized concepts. Results and source code are in our anonymous project page. |
| title | TweedieMix: Improving Multi-Concept Fusion for Diffusion-based Image/Video Generation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2410.05591 |