Research map / Image generation / Editing and style transfer
Attribute-Driven Spontaneous Motion in Unpaired Image Translation
Ruizheng Wu, Xin Tao, Xiaodong Gu, Xiaoyong Shen, Jiaya Jia
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 the spontaneous motion module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Smiling | FID (x1e3), lower is better: 3.224 -> 2.907 | Improved |
| add the spontaneous motion module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Big_Nose | FID (x1e3), lower is better: 6.051 -> 5.137 | Improved |
| add the spontaneous motion module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Pointy_Nose | FID (x1e3), lower is better: 5.682 -> 4.704 | Improved |
| add the spontaneous motion module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Arched_Eyebrow | FID (x1e3), lower is better: 4.667 -> 3.678 | Improved |
| add the spontaneous motion module (ablation) | Ours (spontaneous motion + refinement) · RaFD expression classification | accuracy (%): 88.29 -> 97.85 | Improved |
| add the refinement module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Smiling | FID (x1e3), lower is better: 3.022 -> 2.907 | Improved |
| add the refinement module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Big_Nose | FID (x1e3), lower is better: 6.505 -> 5.137 | Improved |
| add the refinement module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Pointy_Nose | FID (x1e3), lower is better: 5.894 -> 4.704 | Improved |
| add the refinement module (ablation) | Ours (spontaneous motion + refinement) · Face attribute translation: Arched_Eyebrow | FID (x1e3), lower is better: 3.911 -> 3.678 | Improved |
| add the refinement module (ablation) | Ours (spontaneous motion + refinement) · RaFD expression classification | accuracy (%): 90.96 -> 97.85 | Improved |
| Ours (Full) versus StarGAN on classification accuracy | Ours (Full) · RaFD expression classification | accuracy (%): 95.67 -> 97.85 | Improved |
| Ours (Full) versus StarGAN on Big_Nose user preference | Ours (Full) · User study: Big_Nose | ratio selected as best (%): 25.66 -> 52.21 | 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 6 of 6 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
- Learning Temporally and Semantically Consistent Unpaired Video-to-Video Translation through Pseudo-Supervision from Synthetic Optical Flow (AAAI 2022 · similarity 0.85)
- Unpaired Image-to-Image Translation With Shortest Path Regularization (CVPR 2023 · similarity 0.85)
- The Spatially-Correlative Loss for Various Image Translation Tasks (CVPR 2021 · similarity 0.85)
- Unsupervised Image-to-Image Translation with Density Changing Regularization (NeurIPS 2022 · similarity 0.84)
- ReMatching Dynamic Reconstruction Flow (ICLR 2025 · similarity 0.83)
- Revisiting Global Translation Estimation with Feature Tracks (CVPR 2024 · similarity 0.83)
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.
