Research map / Image generation / GANs
Inner Space Preserving Generative Pose Machine
Shuangjun Liu, Sarah Ostadabbas
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 a patch discriminator (cGAN loss) to L1-only training (ablation) | ISP-GPM · SURREAL reposing | texture quality (qualitative) | Improved |
| increase discriminator depth from two to four layers (larger patches) (ablation) | ISP-GPM · SURREAL reposing | image artifacts (qualitative) | Hurt |
| replace max pooling downsampling with convolutional downsampling in hourglass (ablation) | ISP-GPM · SURREAL reposing | texture detail vs color fidelity (qualitative) | Mixed |
| remove discriminator, use only L1 loss (ablation) | ISP-GPM · SURREAL reposing | L1 training loss | Improved |
| use small lambda for L1 weight instead of lambda 100 (ablation) | ISP-GPM · SURREAL reposing | transparency artifacts (qualitative) | Hurt |
Closest papers by meaning
- Intrinsic-Extrinsic Preserved GANs for Unsupervised 3D Pose Transfer (ICCV 2021 · similarity 0.87)
- Deformable GANs for Pose-Based Human Image Generation (CVPR 2018 · similarity 0.86)
- Appearance and Pose-Conditioned Human Image Generation Using Deformable GANs (TPAMI 2019 · similarity 0.85)
- Scene-Aware Generative Network for Human Motion Synthesis (CVPR 2021 · similarity 0.85)
- Multistage Adversarial Losses for Pose-Based Human Image Synthesis (CVPR 2018 · similarity 0.84)
- Can Generative Video Models Help Pose Estimation? (CVPR 2025 · 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.
