Research map / Image generation / Editing and style transfer
MIDMs: Matching Interleaved Diffusion Models for Exemplar-Based Image Translation
Junyoung Seo, Gyuseong Lee, Seokju Cho, Jiyoung Lee, Seungryong Kim
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| interleave recurrent cross-domain matching with diffusion denoising (ablation) | MIDMs · CelebA-HQ | FID: 24.76 -> 15.67 | Improved |
| add confidence masking based on cycle consistency (ablation) | MIDMs · CelebA-HQ | FID: 19.21 -> 15.67 | Improved |
| retain the diffusion U-Net (ablation) | MIDMs · CelebA-HQ | FID: 128.70 -> 15.67 | Improved |
| use a 25% sampling noise level (ablation) | MIDMs · CelebA-HQ | FID: 23.67 (20%) -> 15.67 (25%) | Improved |
| add cycle-consistency loss (ablation) | MIDMs · CelebA-HQ | FID: 16.18 -> 15.67 | Improved |
| add style loss (ablation) | MIDMs · CelebA-HQ | FID: 19.23 -> 15.67 | Improved |
| add perceptual loss (ablation) | MIDMs · CelebA-HQ | FID: 16.51 -> 15.67 | Improved |
| add domain loss (ablation) | MIDMs · CelebA-HQ | FID: 16.68 -> 15.67 | Improved |
| add source loss (ablation) | MIDMs · CelebA-HQ | FID: 72.25 -> 15.67 | Improved |
| increase sampling steps from T=16, N=4 to T=200, N=50 (ablation) | MIDMs · CelebA-HQ | FID: 17.94 -> 15.67 | Improved |
| apply iterative matching and generation using a GAN-based generator (ablation) | CoCosNet · DeepFashion | FID: 14.4 -> 19.05 | Hurt |
| use MIDMs instead of MCL-Net | exemplar-based translation · CelebA-HQ w/ sources | FID: 12.52 -> 15.67 | Hurt |
Later papers that compared with this method
1 later accepted paper reports MIDMs as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- EBDM: Exemplar-guided Image Translation with Brownian-bridge Diffusion Models (ECCV 2024 · better on 3 of 4 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CoCosNet v2 — CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation (CVPR 2021) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- NoiseCLR: A Contrastive Learning Approach for Unsupervised Discovery of Interpretable Directions in Diffusion Models (CVPR 2024 · similarity 0.88)
- Diffusion Model for Dense Matching (ICLR 2024 · similarity 0.88)
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.88)
- I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models (ICML 2025 · similarity 0.87)
- Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion Models (ICLR 2024 · similarity 0.87)
- Simpler Diffusion: 1.5 FID on ImageNet512 with Pixel-space Diffusion (CVPR 2025 · 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.
