Research map / 3D vision / Human pose estimation
Deformation-Aware Unpaired Image Translation for Pose Estimation on Laboratory Animals
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 |
|---|---|---|---|
| use deformation-aware unpaired translation instead of Cycle-GAN-generated training images | pose estimation on Drosophila melanogaster · Drosophila melanogaster | PI-PCK (15 pix): 72.9 -> 83.2 | Helped |
| use deformation-aware unpaired translation instead of Cycle-GAN-generated training images | pose estimation on C. elegans · C. elegans | PI-PCK (5 pix): 45.3 -> 64.0 | Helped |
| use deformation-aware unpaired translation instead of Cycle-GAN-generated training images | pose estimation on D. rerio · D. rerio | PCK (10 pix): 68.7 -> 85.0 | Helped |
| add global deformation (ablation) | full deformation-aware method · Drosophila melanogaster | PI-PCK (15 pix): 79.2 -> 83.2 | Helped |
| add deformation (ablation) | full deformation-aware method · Drosophila melanogaster | PI-PCK (15 pix): 64.9 -> 83.2 | Helped |
| add local deformation (ablation) | full deformation-aware method · Drosophila melanogaster | PI-PCK (15 pix): 57.4 -> 83.2 | Helped |
| predict the deformation as a spatial gradient rather than a vector field directly (ablation) | deformation-aware method · Drosophila melanogaster | PI-PCK (15 pix): 69.1 -> 83.2 | Helped |
| reduce batch size from 128 to 12 (ablation) | deformation-aware method · Drosophila melanogaster | PI-PCK (15 pix): 74.4 -> 83.2 | Helped |
| include the image discriminator D_I and end-to-end training (ablation) | image generation for pose-estimation training · Drosophila melanogaster | image quality | Helped |
| include the image discriminator D_I and end-to-end training (ablation) | image generation for pose-estimation training · C. elegans | image quality | No clear effect |
| use explicit deformation and silhouette modeling instead of Cycle-GAN | image translation · Drosophila melanogaster | SSIM: 0.6543 -> 0.6619 | Helped |
Closest papers by meaning
- From Synthetic to Real: Unsupervised Domain Adaptation for Animal Pose Estimation (CVPR 2021 · similarity 0.89)
- Pose Splatter: A 3D Gaussian Splatting Model for Quantifying Animal Pose and Appearance (NeurIPS 2025 · similarity 0.88)
- Learning From Synthetic Animals (CVPR 2020 · similarity 0.87)
- R^2-Art: Category-Level Articulation Pose Estimation from Single RGB Image via Cascade Render Strategy (AAAI 2025 · similarity 0.85)
- Source-Free and Image-Only Unsupervised Domain Adaptation for Category Level Object Pose Estimation (ICLR 2024 · similarity 0.85)
- PoseGen: Learning to Generate 3D Human Pose Dataset with NeRF (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.