Research map / Image generation / Editing and style transfer
Alleviating Semantics Distortion in Unsupervised Low-Level Image-to-Image Translation via Structure Consistency Constraint
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 SCC to CUT | CUT · GTA → Cityscapes | mean IoU: 0.095 -> 0.11 | Helped |
| Add SCC to CUT | CUT · Cityscapes parsing → image | mean IoU: 0.178 -> 0.182 | Helped |
| Add SCC to GcGAN-rot | GcGAN-rot · Photo → Map | RMSE: 27.98 -> 26.55 | Helped |
| Add SCC to GcGAN-rot | GcGAN-rot · Photo → Map | acc%(δ1): 42.8 -> 44.7 | Helped |
| Add SCC to GcGAN-rot | GcGAN-rot · Photo → Map | acc%(δ2): 64.6 -> 66.5 | Helped |
| Add SCC to CycleGAN | CycleGAN · GTA → Cityscapes | mean IoU: 0.043 -> 0.076 | Helped |
| Add SCC to GAN | GAN · GTA → Cityscapes | mean IoU: 0.068 -> 0.089 | Helped |
| Add SCC to CUT | CUT · Photo → Map | acc%(δ1): 40.1 -> 39.2 | Hurt |
| Add SCC to CUT | CUT · Photo → Map | acc%(δ2): 61.2 -> 60.5 | Hurt |
| Increase SCC weight from 0 to 9 (ablation) | CycleGAN + SCC · GTA → Cityscapes | pixel accuracy: 0.232 -> 0.386 | Helped |
| Increase SCC weight from 7 to 9 (ablation) | CycleGAN + SCC · GTA → Cityscapes | pixel accuracy: 0.382 -> 0.386 | Helped |
| Use an excessively large SCC weight (ablation) | CycleGAN + SCC · GTA → Cityscapes | translation style quality | Hurt |
Closest papers by meaning
- The Spatially-Correlative Loss for Various Image Translation Tasks (CVPR 2021 · similarity 0.88)
- Unsupervised Image-to-Image Translation Using Domain-Specific Variational Information Bound (NeurIPS 2018 · similarity 0.88)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 2024 · similarity 0.86)
- Noise Diffusion for Enhancing Semantic Faithfulness in Text-to-Image Synthesis (CVPR 2025 · similarity 0.86)
- Leveraging Unpaired Data for Vision-Language Generative Models via Cycle Consistency (ICLR 2024 · similarity 0.86)
- MARS: Mixture of Auto-Regressive Models for Fine-grained Text-to-image Synthesis (AAAI 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.