Research map / Image generation / Editing and style transfer
LEED: Label-Free Expression Editing via Disentanglement
Rongliang Wu, Shijian Lu
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 |
|---|---|---|---|
| replace GANimation with LEED | GANimation · RaFD | FID: 45.55 -> 38.20 | Improved |
| replace GANimation with LEED | GANimation · CFEED | FID: 29.07 -> 23.60 | Improved |
| replace GANimation with LEED | GANimation · RaFD | SSIM: 0.8686 -> 0.8833 | Improved |
| replace GANimation with LEED | GANimation · CFEED | expression accuracy (R+G): 84.42 -> 90.06 | Improved |
| add LEED-edited images to classifier training | real-image classifier · RaFD | expression accuracy: 92.21 -> 93.25 | Improved |
| add LEED-edited images to classifier training | real-image classifier · CFEED | expression accuracy: 88.23 -> 90.06 | Improved |
| add GANimation-edited images to classifier training | real-image classifier · CFEED | expression accuracy: 88.23 -> 84.42 | Hurt |
| add StarGAN-edited images to classifier training | real-image classifier · RaFD | expression accuracy: 92.21 -> 88.48 | Hurt |
| replace GANimation with LEED | GANimation · RaFD | AMT Which's More Real: 15.29 -> 77.65 | Improved |
| replace GANimation with LEED | GANimation · CFEED | AMT Real or Fake: 31.87 -> 70.63 | Improved |
| add mutual expression information loss (ablation) | LEED · RaFD | qualitative editing quality | Improved |
| add siamese loss (ablation) | LEED · RaFD | qualitative expression intensity | 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 12 of 12 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
- FED-Bench: A Cross-Granular Benchmark for Disentangled Evaluation of Facial Expression Editing (ECCV 2026 · similarity 0.89)
- Face2Diffusion for Fast and Editable Face Personalization (CVPR 2024 · similarity 0.85)
- SDGAN: Disentangling Semantic Manipulation for Facial Attribute Editing (AAAI 2024 · similarity 0.85)
- AUEditNet: Dual-Branch Facial Action Unit Intensity Manipulation with Implicit Disentanglement (CVPR 2024 · similarity 0.84)
- Lifting Scheme-Based Implicit Disentanglement of Emotion-Related Facial Dynamics in the Wild (AAAI 2025 · similarity 0.84)
- Navigating Label Ambiguity for Facial Expression Recognition in the Wild (AAAI 2025 · similarity 0.84)
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.
