Research map / Image generation / Editing and style transfer
Masked and Adaptive 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 MATEBIT instead of DynaST | exemplar-based translation on CelebA-HQ · CelebA-HQ | FID: 12.0 -> 11.5 | Helped |
| use MATEBIT instead of the strongest listed baseline | exemplar-based translation on Metfaces · Metfaces | FID: 23.3 -> 26.0 | Hurt |
| use MATEBIT instead of DynaST | exemplar-based translation on Cartoon · Cartoon | FID: 62.8 -> 64.4 | Hurt |
| use MATEBIT instead of CoCosNet-v2 | exemplar-based translation on Ukiyo-e · Ukiyo-e | FID: 32.1 -> 30.3 | Helped |
| use MATEBIT instead of MCL-Net | exemplar-based translation on AAHQ · AAHQ | FID: 64.4 -> 56.0 | Helped |
| use MATEBIT instead of DynaST | exemplar-based translation on DeepFashion · DeepFashion | FID: 8.4 -> 8.2 | Helped |
| add MAT (ablation) | MATEBIT · Metfaces | FID: 27.7 -> 26.0 | Helped |
| add ReLU masking in MAT (ablation) | MATEBIT · Metfaces | FID: 30.1 -> 26.0 | Helped |
| use MAT instead of three-layer full correspondence learning (ablation) | MATEBIT · Metfaces | FID: 34.1 -> 26.0 | Helped |
| use AdaConv instead of FFN (ablation) | MATEBIT · Metfaces | FID: 34.3 -> 26.0 | Helped |
| add contrastive style loss (ablation) | MATEBIT · Metfaces | Texture: 0.874 -> 0.938 | Helped |
| use the proposed style loss instead of CAST loss (ablation) | MATEBIT · Metfaces | FID: 31.0 -> 26.0 | Helped |
Closest papers by meaning
- Cross-Domain Correspondence Learning for Exemplar-Based Image Translation (CVPR 2020 · similarity 0.92)
- Exemplar Guided Unsupervised Image-to-Image Translation with Semantic Consistency (ICLR 2019 · similarity 0.90)
- Marginal Contrastive Correspondence for Guided Image Generation (CVPR 2022 · similarity 0.89)
- Encoding in Style: A StyleGAN Encoder for Image-to-Image Translation (CVPR 2021 · similarity 0.89)
- Learning to Manipulate Artistic Images (AAAI 2024 · similarity 0.87)
- ENAT: Rethinking Spatial-temporal Interactions in Token-based Image Synthesis (NeurIPS 2024 · similarity 0.85)
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.