Research map / Image generation / Editing and style transfer
Informative Sample Mining Network for Multi-Domain Image-to-Image Translation
Jie Cao, Huaibo Huang, Yi Li, Ran He, Zhenan Sun
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 |
|---|---|---|---|
| add Adversarial Importance Weighting (AIW) to 2-hop MST (ablation) | INIT · CelebA-HQ facial attribute transfer | FID: 18.91 -> 11.16 | Improved |
| add AIW to 2-hop MST (ablation) | INIT · CelebA-HQ facial attribute transfer | mean attribute accuracy: 92.86 -> 95.08 | Improved |
| add AIW to the plain conditional GAN (ablation) | conditional GAN · CelebA-HQ facial attribute transfer | FID: 21.38 -> 14.52 | Improved |
| add multi-hop sample training (2-hop) to AIW (ablation) | INIT · CelebA-HQ facial attribute transfer | FID: 14.52 -> 11.16 | Improved |
| add multi-hop sample training (2-hop) to AIW (ablation) | INIT · CelebA-HQ facial attribute transfer | mean attribute accuracy: 93.78 -> 95.08 | Improved |
| add both AIW and 2-hop MST to conditional GAN (ablation) | conditional GAN · CelebA-HQ facial attribute transfer | FID: 21.38 -> 11.16 | Improved |
| increase hop number from 2 to 3 (ablation) | INIT · CelebA-HQ facial attribute transfer | FID: 11.16 -> 11.23 | No clear effect |
| increase hop number from 2 to 3 (ablation) | INIT · CelebA-HQ facial attribute transfer | mean attribute accuracy: 95.08 -> 94.98 | Hurt |
| use INIT instead of STGAN | facial attribute transfer · CelebA-HQ | FID: 15.94 -> 11.16 | Improved |
| use INIT instead of STGAN | season transfer · Yosemite winter | FID: 50.37 -> 44.91 | Improved |
| use INIT instead of AttGAN | edge-to-photo · Edge2Photo shoes, photo FID | FID: 29.01 -> 28.44 | Improved |
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.
- StarGAN — StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation (CVPR 2018) · better on 7 of 7 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
- Batch Weight for Domain Adaptation With Mass Shift (ICCV 2019 · similarity 0.87)
- Theoretical Analysis of Image-to-Image Translation with Adversarial Learning (ICML 2018 · similarity 0.86)
- Discriminative Region Proposal Adversarial Networks for High-Quality Image-to-Image Translation (ECCV 2018 · similarity 0.86)
- Discriminative Region Proposal Adversarial Network for High-Quality Image-to-Image Translation (IJCV 2019 · similarity 0.86)
- TraVeLGAN: Image-To-Image Translation by Transformation Vector Learning (CVPR 2019 · similarity 0.86)
- Improving the Training of Data-Efficient GANs via Quality Aware Dynamic Discriminator Rejection Sampling (CVPR 2025 · similarity 0.86)
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.
