Research map / 3D vision / Face and shape
3D-Aware Semantic-Guided Generative Model for Human Synthesis
Jichao Zhang, Enver Sangineto, Hao Tang, Aliaksandr Siarohin, Zhun Zhong, Nicu Sebe, Wei Wang
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 |
|---|---|---|---|
| use semantic-mask-guided 3D-SGAN instead of GIRAFFE | 3D-aware human GAN · DeepFashion Random | FID: 42.73 -> 8.240 | Improved |
| use 3D-SGAN instead of GIRAFFE | 3D-aware human GAN · VITON | FID: 26.750 -> 14.060 | Improved |
| use 3D-SGAN instead of ShadeGAN | 3D-aware human GAN · DeepFashion Rotation | aMP: 89.0 -> 81.0 | Hurt |
| use 3D-SGAN instead of CIPS-3D | 3D-aware human GAN · DeepFashion Texture | FID: 36.16 -> 60.63 | Hurt |
| add semantic consistency loss l_ss (ablation) | 3D-SGAN · DeepFashion consistency | L1 silhouette distance: 5.2489 -> 3.9614 | Improved |
| add pose-semantic consistency loss l_ps (ablation) | 3D-SGAN · DeepFashion pose consistency | LPIPS diversity: 0.1132 -> 0.0393 | Improved |
| use GNeRF 3D generator instead of vanilla 2D GAN (ablation) | 3D-SGAN · DeepFashion | FID: 13.24 -> 8.240 | Improved |
| add semantic masks as intermediate representation (ablation) | 3D-SGAN · DeepFashion | FID: 66.79 -> 8.240 | Improved |
| add VAE-conditioned texture generator (ablation) | 3D-SGAN · DeepFashion | FID: 14.35 -> 8.240 | Improved |
| train G3D and Gt jointly instead of separately (ablation) | 3D-SGAN · DeepFashion | FID | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- GIRAFFE — GIRAFFE: Representing Scenes As Compositional Generative Neural Feature Fields (CVPR 2021) · better on 6 of 6 shared results
- GRAF — GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis (NeurIPS 2020) · better on 6 of 6 shared results
- pi-GAN — Pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis (CVPR 2021) · better on 4 of 4 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
- SemanticHuman-HD: High Resolution Semantic disentangled 3D Human Generation (ECCV 2024 · similarity 0.90)
- Generative Neural Articulated Radiance Fields (NeurIPS 2022 · similarity 0.90)
- GSNeRF: Generalizable Semantic Neural Radiance Fields with Enhanced 3D Scene Understanding (CVPR 2024 · similarity 0.89)
- Semantic 3D-Aware Portrait Synthesis and Manipulation Based on Compositional Neural Radiance Field (AAAI 2023 · similarity 0.88)
- GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis (NeurIPS 2020 · similarity 0.88)
- GRF: Learning a General Radiance Field for 3D Representation and Rendering (ICCV 2021 · similarity 0.88)
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.
