Research map / Image generation / Editing and style transfer
Unbalanced Feature Transport for Exemplar-Based Image Translation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use UNITE instead of CoCosNet | exemplar-based translation · ADE20K | FID: 26.40 -> 25.15 | Helped |
| use UNITE instead of CoCosNet | exemplar-based translation · COCO-Stuff | FID: 35.23 -> 33.65 | Helped |
| use UNITE instead of CoCosNet | exemplar-based translation · DeepFashion | FID: 14.40 -> 13.08 | Helped |
| use UNITE instead of CoCosNet | exemplar-based translation · CelebA-HQ | FID: 14.30 -> 13.15 | Helped |
| add cosine feature alignment (ablation) | SPADE · CelebA-HQ | FID: 31.50 -> 16.32 | Helped |
| replace cosine matching with classical optimal transport (ablation) | SPADE+COS · CelebA-HQ | FID: 16.32 -> 17.87 | Hurt |
| replace classical optimal transport with unbalanced optimal transport (ablation) | SPADE+OT · CelebA-HQ | FID: 17.87 -> 14.02 | Helped |
| replace SPADE with SEACE (ablation) | SPADE+UOT · CelebA-HQ | FID: 14.02 -> 13.46 | Helped |
| add multi-stage feature transport (ablation) | SEACE+UOT · CelebA-HQ | FID: 13.46 -> 13.15 | Helped |
| add multi-stage feature transport (ablation) | SEACE+UOT · CelebA-HQ | SWD: 15.12 -> 14.91 | Helped |
| add multi-stage feature transport (ablation) | SEACE+UOT · CelebA-HQ | LPIPS: 0.208 -> 0.213 | Helped |
| use UNITE instead of CoCosNet | exemplar-based translation · DeepFashion | LPIPS: 0.296 -> 0.278 | Hurt |
Closest papers by meaning
- Exemplar Guided Unsupervised Image-to-Image Translation with Semantic Consistency (ICLR 2019 · similarity 0.89)
- TraVeLGAN: Image-To-Image Translation by Transformation Vector Learning (CVPR 2019 · similarity 0.89)
- Cross-Domain Correspondence Learning for Exemplar-Based Image Translation (CVPR 2020 · similarity 0.89)
- Masked and Adaptive Transformer for Exemplar Based Image Translation (CVPR 2023 · similarity 0.87)
- Unsupervised Image-to-Image Translation With Generative Prior (CVPR 2022 · similarity 0.87)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.