CTRLorALTer: Conditional LoRAdapter for Efficient 0-Shot Control & Altering of T2I Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913536693764096 |
|---|---|
| author | Stracke, Nick Baumann, Stefan Andreas Susskind, Joshua M. Bautista, Miguel Angel Ommer, Björn |
| author_facet | Stracke, Nick Baumann, Stefan Andreas Susskind, Joshua M. Bautista, Miguel Angel Ommer, Björn |
| contents | Text-to-image generative models have become a prominent and powerful tool that excels at generating high-resolution realistic images. However, guiding the generative process of these models to consider detailed forms of conditioning reflecting style and/or structure information remains an open problem. In this paper, we present LoRAdapter, an approach that unifies both style and structure conditioning under the same formulation using a novel conditional LoRA block that enables zero-shot control. LoRAdapter is an efficient, powerful, and architecture-agnostic approach to condition text-to-image diffusion models, which enables fine-grained control conditioning during generation and outperforms recent state-of-the-art approaches. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_07913 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | CTRLorALTer: Conditional LoRAdapter for Efficient 0-Shot Control & Altering of T2I Models Stracke, Nick Baumann, Stefan Andreas Susskind, Joshua M. Bautista, Miguel Angel Ommer, Björn Computer Vision and Pattern Recognition Text-to-image generative models have become a prominent and powerful tool that excels at generating high-resolution realistic images. However, guiding the generative process of these models to consider detailed forms of conditioning reflecting style and/or structure information remains an open problem. In this paper, we present LoRAdapter, an approach that unifies both style and structure conditioning under the same formulation using a novel conditional LoRA block that enables zero-shot control. LoRAdapter is an efficient, powerful, and architecture-agnostic approach to condition text-to-image diffusion models, which enables fine-grained control conditioning during generation and outperforms recent state-of-the-art approaches. |
| title | CTRLorALTer: Conditional LoRAdapter for Efficient 0-Shot Control & Altering of T2I Models |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2405.07913 |