Research map / 3D vision / Face and shape
3DFaceGAN: Adversarial Nets for 3D Face Representation, Generation, and Translation
Stylianos Moschoglou, Stylianos Ploumpis, Mihalis A. Nicolaou, Athanasios Papaioannou, Stefanos P. Zafeiriou
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 discriminator pre-training before adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face representation ablation | AUC: 0.704 -> 0.741 | Improved |
| add discriminator pre-training before adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face representation ablation | failure rate (%): 3.15e-6 -> 1.42e-7 | Improved |
| freeze decoder weights of G and D during adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face representation ablation | AUC: 0.736 -> 0.741 | Improved |
| freeze decoder weights of G and D during adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face representation ablation | failure rate (%): 2.62e-7 -> 1.42e-7 | Improved |
| add adversarial training with pre-trained autoencoder discriminator over plain AE (ablation) | 3DFaceGAN · Hi-Lo 3D face representation ablation | AUC: 0.697 -> 0.741 | Improved |
| add discriminator pre-training before adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face translation ablation | 3DRMSE AUC: 0.794 -> 0.827 | Improved |
| freeze decoder weights of G and D during adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face translation ablation | 3DRMSE AUC: 0.819 -> 0.827 | Improved |
| freeze decoder weights of G and D during adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face translation ablation | failure rate (%): 8.70e-6 -> 5.49e-6 | Improved |
| add adversarial training with pre-trained autoencoder discriminator over denoising AE (ablation) | 3DFaceGAN · Hi-Lo 3D face translation ablation | 3DRMSE AUC: 0.758 -> 0.827 | Improved |
| add discriminator pre-training before adversarial training (ablation) | 3DFaceGAN · Hi-Lo 3D face translation ablation | failure rate (%): 1.38e-5 -> 5.49e-6 | Improved |
Later papers that compared with this method
1 later accepted paper reports 3DFaceGAN as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Unpaired Multi-domain Attribute Translation of 3D Facial Shapes with a Square and Symmetric Geometric Map (ICCV 2023 · better on 2 of 2 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
- GANFIT: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction (CVPR 2019 · similarity 0.92)
- Synthesizing Coupled 3D Face Modalities by Trunk-Branch Generative Adversarial Networks (ECCV 2020 · similarity 0.92)
- Next3D: Generative Neural Texture Rasterization for 3D-Aware Head Avatars (CVPR 2023 · similarity 0.91)
- Fast-GANFIT: Generative Adversarial Network for High Fidelity 3D Face Reconstruction (TPAMI 2021 · similarity 0.91)
- A Decoupled 3D Facial Shape Model by Adversarial Training (ICCV 2019 · similarity 0.91)
- A 3D GAN for Improved Large-Pose Facial Recognition (CVPR 2021 · similarity 0.91)
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.
