Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.07221 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911368742961152 |
|---|---|
| author | Ryu, Jongwon Park, Joonhyung Han, Jaeho Kim, Yeong-Seok Kim, Hye-rin Yoon, Sunjae Kim, Junyeong |
| author_facet | Ryu, Jongwon Park, Joonhyung Han, Jaeho Kim, Yeong-Seok Kim, Hye-rin Yoon, Sunjae Kim, Junyeong |
| contents | Multi-domain image-to-image translation re quires grounding semantic differences ex pressed in natural language prompts into corresponding visual transformations, while preserving unrelated structural and seman tic content. Existing methods struggle to maintain structural integrity and provide fine grained, attribute-specific control, especially when multiple domains are involved. We propose LACE (Language-grounded Attribute Controllable Translation), built on two compo nents: (1) a GLIP-Adapter that fuses global semantics with local structural features to pre serve consistency, and (2) a Multi-Domain Control Guidance mechanism that explicitly grounds the semantic delta between source and target prompts into per-attribute translation vec tors, aligning linguistic semantics with domain level visual changes. Together, these modules enable compositional multi-domain control with independent strength modulation for each attribute. Experiments on CelebA(Dialog) and BDD100K demonstrate that LACE achieves high visual fidelity, structural preservation, and interpretable domain-specific control, surpass ing prior baselines. This positions LACE as a cross-modal content generation framework bridging language semantics and controllable visual translation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_07221 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Language-Grounded Multi-Domain Image Translation via Semantic Difference Guidance Ryu, Jongwon Park, Joonhyung Han, Jaeho Kim, Yeong-Seok Kim, Hye-rin Yoon, Sunjae Kim, Junyeong Computer Vision and Pattern Recognition Multi-domain image-to-image translation re quires grounding semantic differences ex pressed in natural language prompts into corresponding visual transformations, while preserving unrelated structural and seman tic content. Existing methods struggle to maintain structural integrity and provide fine grained, attribute-specific control, especially when multiple domains are involved. We propose LACE (Language-grounded Attribute Controllable Translation), built on two compo nents: (1) a GLIP-Adapter that fuses global semantics with local structural features to pre serve consistency, and (2) a Multi-Domain Control Guidance mechanism that explicitly grounds the semantic delta between source and target prompts into per-attribute translation vec tors, aligning linguistic semantics with domain level visual changes. Together, these modules enable compositional multi-domain control with independent strength modulation for each attribute. Experiments on CelebA(Dialog) and BDD100K demonstrate that LACE achieves high visual fidelity, structural preservation, and interpretable domain-specific control, surpass ing prior baselines. This positions LACE as a cross-modal content generation framework bridging language semantics and controllable visual translation. |
| title | Language-Grounded Multi-Domain Image Translation via Semantic Difference Guidance |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.07221 |