Controlling Text-to-Image Diffusion by Orthogonal Finetuning
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2023
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917613204930560 |
|---|---|
| author | Qiu, Zeju Liu, Weiyang Feng, Haiwen Xue, Yuxuan Feng, Yao Liu, Zhen Zhang, Dan Weller, Adrian Schölkopf, Bernhard |
| author_facet | Qiu, Zeju Liu, Weiyang Feng, Haiwen Xue, Yuxuan Feng, Yao Liu, Zhen Zhang, Dan Weller, Adrian Schölkopf, Bernhard |
| contents | Large text-to-image diffusion models have impressive capabilities in generating photorealistic images from text prompts. How to effectively guide or control these powerful models to perform different downstream tasks becomes an important open problem. To tackle this challenge, we introduce a principled finetuning method -- Orthogonal Finetuning (OFT), for adapting text-to-image diffusion models to downstream tasks. Unlike existing methods, OFT can provably preserve hyperspherical energy which characterizes the pairwise neuron relationship on the unit hypersphere. We find that this property is crucial for preserving the semantic generation ability of text-to-image diffusion models. To improve finetuning stability, we further propose Constrained Orthogonal Finetuning (COFT) which imposes an additional radius constraint to the hypersphere. Specifically, we consider two important finetuning text-to-image tasks: subject-driven generation where the goal is to generate subject-specific images given a few images of a subject and a text prompt, and controllable generation where the goal is to enable the model to take in additional control signals. We empirically show that our OFT framework outperforms existing methods in generation quality and convergence speed. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2306_07280 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Controlling Text-to-Image Diffusion by Orthogonal Finetuning Qiu, Zeju Liu, Weiyang Feng, Haiwen Xue, Yuxuan Feng, Yao Liu, Zhen Zhang, Dan Weller, Adrian Schölkopf, Bernhard Computer Vision and Pattern Recognition Artificial Intelligence Graphics Machine Learning Large text-to-image diffusion models have impressive capabilities in generating photorealistic images from text prompts. How to effectively guide or control these powerful models to perform different downstream tasks becomes an important open problem. To tackle this challenge, we introduce a principled finetuning method -- Orthogonal Finetuning (OFT), for adapting text-to-image diffusion models to downstream tasks. Unlike existing methods, OFT can provably preserve hyperspherical energy which characterizes the pairwise neuron relationship on the unit hypersphere. We find that this property is crucial for preserving the semantic generation ability of text-to-image diffusion models. To improve finetuning stability, we further propose Constrained Orthogonal Finetuning (COFT) which imposes an additional radius constraint to the hypersphere. Specifically, we consider two important finetuning text-to-image tasks: subject-driven generation where the goal is to generate subject-specific images given a few images of a subject and a text prompt, and controllable generation where the goal is to enable the model to take in additional control signals. We empirically show that our OFT framework outperforms existing methods in generation quality and convergence speed. |
| title | Controlling Text-to-Image Diffusion by Orthogonal Finetuning |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Graphics Machine Learning |
| url | https://arxiv.org/abs/2306.07280 |