Research map / Image generation / Editing and style transfer
Exploring Patch-Wise Semantic Relation for Contrastive Learning in Image-to-Image Translation Tasks
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 |
|---|---|---|---|
| Add semantic relation consistency (SRC) to DCE (ablation) | CUT · Horse→Zebra | FID: 41.2 -> 36.2 | Helped |
| Add hard negative mining to DCE + SRC (ablation) | CUT · Horse→Zebra | FID: 36.2 -> 34.4 | Helped |
| Replace InfoNCE with DCE (ablation) | CUT · Horse→Zebra | FID: 45.5 -> 41.2 | Helped |
| Add SRC to InfoNCE (ablation) | CUT · Horse→Zebra | FID: 45.5 -> 42.7 | Helped |
| Add hard negative mining to InfoNCE + SRC (ablation) | CUT · Horse→Zebra | FID: 42.7 -> 40.5 | Helped |
| Add semantic relation consistency (SRC) to DCE (ablation) | CUT · Cityscapes label→image | FID: 52.0 -> 49.9 | Helped |
| Add hard negative mining to DCE + SRC (ablation) | CUT · Cityscapes label→image | FID: 49.9 -> 46.4 | Helped |
| Replace InfoNCE with DCE (ablation) | CUT · Cityscapes label→image | FID: 56.4 -> 52.0 | Helped |
| Add hard negative mining to DCE without SRC (ablation) | CUT · Cityscapes label→image | mAP: 27.8 -> 27.3 | Hurt |
| Add semantic relation consistency and decoupled contrastive loss | StarGANv2 · Seasons latent | LPIPS: 0.413 -> 0.496 | Helped |
| Add semantic relation consistency and decoupled contrastive loss | StarGANv2 · Weather reference | FID: 64.20 -> 56.91 | Helped |
| Add semantic relation distillation | Fast GAN Compression · Cityscapes compression | mIOU: 41.42 -> 43.06 | Helped |
Closest papers by meaning
- Patch-Wise Graph Contrastive Learning for Image Translation (AAAI 2024 · similarity 0.89)
- Alleviating Semantics Distortion in Unsupervised Low-Level Image-to-Image Translation via Structure Consistency Constraint (CVPR 2022 · similarity 0.88)
- CoCosNet v2: Full-Resolution Correspondence Learning for Image Translation (CVPR 2021 · similarity 0.87)
- Self-Supervised Dense Consistency Regularization for Image-to-Image Translation (CVPR 2022 · similarity 0.87)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.86)
- SynthRGB-T: Language-Vision Guided Image Translation for Diversity Synthesis (CVPR 2026 · 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.