Research map / Video / Video segmentation and tracking
Deformable Generator Networks: Unsupervised Disentanglement of Appearance and Geometry
Xianglei Xing, Ruiqi Gao, Tian Han, Song Chun Zhu, Ying Wu
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 deformable (geometric + appearance) factorization vs VAE | VAE · Multi-Pie all views | MSRE: 89.02 -> 76.52 | Improved |
| add deformable factorization vs ABP | ABP · Multi-Pie all views | MSRE: 94.66 -> 76.52 | Improved |
| deformable generator at near-frontal view (0 deg) | VAE and ABP · Multi-Pie 0 deg | MSRE: 48.78 -> 46.10 | No clear effect |
| use dynamic deformable fields instead of geometric latent factors as features | dglf · MUG | accuracy (%): 85.5 -> 92.0 | Improved |
| use dynamic geometric latent factors instead of MoCoGAN motion factors | MoCoGAN · MUG | accuracy (%): 83.1 -> 85.5 | Improved |
| use dynamic deformable fields instead of sparse-landmark methods | UPSM · MUG | accuracy (%): 87.7 -> 92.0 | Improved |
| add adversarial loss (non-saturating + R1) to MLE loss | Deformable generator · FFHQ 256x256 | reconstruction sharpness | Improved |
| learned deformation fields for landmark regression vs MTCNN | MTCNN · MAFL | mean error: 5.39 -> 5.18 | 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.
- MoCoGAN — MoCoGAN: Decomposing Motion and Content for Video Generation (CVPR 2018) · better on 1 of 1 shared result
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
- Unsupervised Disentangling of Appearance and Geometry by Deformable Generator Network (CVPR 2019 · similarity 0.97)
- Disentangling Geometric Deformation Spaces in Generative Latent Shape Models (IJCV 2023 · similarity 0.88)
- Disentangled Pose and Appearance Guidance for Multi-Pose Generation (CVPR 2025 · similarity 0.85)
- Generative Perception of Shape and Material from Differential Motion (NeurIPS 2025 · similarity 0.84)
- Shadows Don't Lie and Lines Can't Bend! Generative Models don't know Projective Geometry...for now (CVPR 2024 · similarity 0.84)
- Disentangled 3D Scene Generation with Layout Learning (ICML 2024 · 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.
