Research map / Image generation / Editing and style transfer
CFFT-GAN: Cross-Domain Feature Fusion Transformer for Exemplar-Based Image Translation
Read the paperCheck 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 CFFT-GAN instead of CoCosNet v2 | Exemplar-based image translation · Deepfashion | FID: 12.2 -> 8.8 | Helped |
| Use CFFT-GAN instead of CoCosNet v2 | Exemplar-based image translation · CelebA-HQ | FID: 16.2 -> 12.4 | Helped |
| Use CFFT-GAN instead of CoCosNet v2 | Exemplar-based image translation · MetFaces | FID: 23.3 -> 22.6 | Helped |
| Use CFFT-GAN instead of CoCosNet v2 | Exemplar-based image translation · ADE20K | FID: 25.2 -> 25.4 | Hurt |
| Use CFFT-GAN instead of CoCosNet v2 | Exemplar-based image translation · Deepfashion | LPIPS: 0.2245 -> 0.2070 | Helped |
| Use CFFT-GAN instead of CoCosNet v2 | Exemplar-based image translation · ADE20K | SWD: 9.9 -> 10.1 | Hurt |
| Replace CFFT with a Correlation Matrix (ablation) | CFFT-GAN · Deepfashion | FID: 8.8 -> 14.3 | Hurt |
| Replace CFFT with AdaIN (ablation) | CFFT-GAN · Deepfashion | FID: 8.8 -> 16.3 | Hurt |
| Add Hiformer to CFFT (ablation) | CFFT-GAN · Deepfashion | FID: 11.3 -> 8.8 | Helped |
| Reduce CFFT feature scale from 64×64 to 32×32 (ablation) | CFFT-GAN · Deepfashion | FID: 8.8 -> 9.5 | Hurt |
Closest papers by meaning
- Masked and Adaptive Transformer for Exemplar Based Image Translation (CVPR 2023 · similarity 0.88)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.84)
- Exact Fusion via Feature Distribution Matching for Few-shot Image Generation (CVPR 2024 · similarity 0.84)
- Puff-Net: Efficient Style Transfer with Pure Content and Style Feature Fusion Network (CVPR 2024 · similarity 0.83)
- Learning to Manipulate Artistic Images (AAAI 2024 · similarity 0.83)
- E$^2$GAN: Efficient Training of Efficient GANs for Image-to-Image Translation (ICML 2024 · similarity 0.83)
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.