Research map / Image generation / Editing and style transfer
Maximum Spatial Perturbation Consistency for Unpaired Image-to-Image Translation
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 maximum spatial perturbation consistency and spatial alignment | MSPC image translation · Cityscapes→Parsing | pixAcc: 0.595 -> 0.740 | Helped |
| Add maximum spatial perturbation consistency and spatial alignment | MSPC image translation · Front Face→Profile | FID: 52.95 -> 37.01 | Helped |
| Add maximum spatial perturbation consistency and spatial alignment | MSPC image translation · Parsing→Cityscapes | mAP: 0.172 -> 0.156 | Hurt |
| Add maximum spatial perturbation consistency and spatial alignment | MSPC image translation · Aerial Photograph→Map | RMSE: 32.70 -> 32.97 | Hurt |
| Vary perturbation scale from a=1 to a=3 (ablation) | MSPC · Front Face→Profile | FID: 42.19 -> 37.01 | Helped |
| Increase perturbation scale from a=3 to a=5 (ablation) | MSPC · Front Face→Profile | FID: 37.01 -> 38.72 | Hurt |
| Increase perturbation scale from a=5 to a=8 (ablation) | MSPC · Front Face→Profile | FID: 38.72 -> 60.21 | Hurt |
| Use random spatial perturbation instead of learned maximum perturbation (ablation) | MSPC · Front Face→Profile | FID: 37.01 -> 67.33 | Hurt |
| Add spatial alignment branch (ablation) | MSPC · Front Face→Profile | FID: 53.41± 4.83 -> 38.61± 2.57 | Helped |
| Apply spatial transformation T to source and target distributions (ablation) | Front Face→Profile data · Front Face→Profile | FID: 112.69 -> 65.81 | Helped |
| Combine translation and spatial alignment rather than translation alone (ablation) | MSPC · Front Face→Profile | FID: 37.01 -> 30.85 | Helped |
Closest papers by meaning
- Domain Adaptive Image-to-Image Translation (CVPR 2020 · similarity 0.89)
- The Spatially-Correlative Loss for Various Image Translation Tasks (CVPR 2021 · similarity 0.88)
- Alleviating Semantics Distortion in Unsupervised Low-Level Image-to-Image Translation via Structure Consistency Constraint (CVPR 2022 · similarity 0.87)
- TraVeLGAN: Image-To-Image Translation by Transformation Vector Learning (CVPR 2019 · similarity 0.87)
- Homomorphic Latent Space Interpolation for Unpaired Image-To-Image Translation (CVPR 2019 · similarity 0.87)
- Rethinking the Paradigm of Content Constraints in Unpaired Image-to-Image Translation (AAAI 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.