| HiFace: High-Fidelity 3D Face Reconstruction by Learning Static and Dynamic Details | | 1.275 | 1.036 (±0.280) | 1.450 (±0.413) | 1.324 (±0.334) | 1.291 (±0.362) | HiFace-f | 2023-03-20 |
| HiFace: High-Fidelity 3D Face Reconstruction by Learning Static and Dynamic Details | | 1.297 | 1.054 (±0.317) | 1.461 (±0.430) | 1.331 (±0.347) | 1.342 (±0.384) | HiFace-c | 2023-03-20 |
| FFHQ-UV: Normalized Facial UV-Texture Dataset for 3D Face Reconstruction | ✓ Link | 1.399 | 1.681 | 1.339 | 1.631 | 0.943 | FFHQ-UV | 2022-11-25 |
| 3D Face Reconstruction with the Geometric Guidance of Facial Part Segmentation | ✓ Link | 1.435 | 1.584 (±0.308) | 1.237 (±0.375) | 1.809 (±0.394) | 1.110 (±0.328) | 3DDFA-v3 | 2023-12-01 |
| MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shading | | 1.500 | 1.499 (±0.366) | 1.424 (±0.462) | 1.950 (±0.559) | 1.128 (±0.303) | MoSAR | 2023-12-20 |
| A Hierarchical Representation Network for Accurate and Detailed Face Reconstruction from In-The-Wild Images | ✓ Link | 1.537 | 1.722 (±0.330) | 1.357 (±0.523) | 1.995 (±0.476) | 1.072 (±0.333) | HRN | 2023-02-28 |
| Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Set | ✓ Link | 1.657 | 1.719 (±0.354) | 1.368 (±0.439) | 2.015 (±0.449) | 1.528 (±0.501) | Deep3D | 2019-03-20 |
| Towards Realistic Generative 3D Face Models | ✓ Link | 1.746 | 1.656 (±0.374) | 2.087 (±0.839) | 2.102 (±0.512) | 1.141 (±0.303) | AlbedoGAN | 2023-04-24 |
| Self-Supervised Monocular 3D Face Reconstruction by Occlusion-Aware Multi-view Geometry Consistency | ✓ Link | 1.787 | 1.827 (±0.383) | 1.409 (±0.418) | 2.248 (±0.508) | 1.665 (±0.644) | MGCNet | 2020-07-24 |
| GANFIT: Generative Adversarial Network Fitting for High Fidelity 3D Face Reconstruction | ✓ Link | 1.868 | 1.928 (±0.490) | 1.812 (±0.544) | 2.402 (±0.545) | 1.329 (±0.504) | GANFit | 2019-02-15 |
| A Perceptual Shape Loss for Monocular 3D Face Reconstruction | | 1.882 | 1.708 (±0.349) | 1.876 (±0.563) | 2.350 (±0.551) | 1.593 (±0.540) | PSL | 2023-10-30 |
| SADRNet: Self-Aligned Dual Face Regression Networks for Robust 3D Dense Face Alignment and Reconstruction | ✓ Link | 1.913 | 1.791 (±0.542) | 1.591 (±0.488) | 2.413 (±0.537) | 1.856 (±0.701) | SADRNet | 2021-06-06 |
| Towards Fast, Accurate and Stable 3D Dense Face Alignment | ✓ Link | 1.926 | 1.903 (±0.517) | 1.597 (±0.478) | 2.447 (±0.647) | 1.757 (±0.642) | 3DDFA-v2 | 2020-09-21 |
| Learning an Animatable Detailed 3D Face Model from In-The-Wild Images | ✓ Link | 2.010 | 1.697 (±0.355) | 2.516 (±0.839) | 2.394 (±0.576) | 1.479 (±0.535) | DECA-c | 2020-12-07 |
| Joint 3D Face Reconstruction and Dense Alignment with Position Map Regression Network | ✓ Link | 2.013 | 1.923 (±0.518) | 1.838 (±0.637) | 2.429 (±0.588) | 1.863 (±0.698) | PRNet | 2018-03-21 |
| Self-Supervised 3D Face Reconstruction via Conditional Estimation | | 2.017 | 2.779 (±0.835) | 1.448 (±0.406) | 2.384 (±0.578) | 1.456 (±0.485) | CEST | 2021-10-10 |
| Synergy between 3DMM and 3D Landmarks for Accurate 3D Facial Geometry | ✓ Link | 2.021 | 2.026 (±0.532) | 1.731 (±0.502) | 2.679 (±0.741) | 1.647 (±0.622) | SynergyNet | 2021-10-19 |
| EMOCA: Emotion Driven Monocular Face Capture and Animation | ✓ Link | 2.103 | 1.868 (±0.387) | 2.679 (±1.112) | 2.426 (±0.641) | 1.438 (±0.501) | EMOCA-c | 2022-04-24 |
| Towards Metrical Reconstruction of Human Faces | ✓ Link | 2.134 | 1.585 (±0.325) | 3.478 (±1.204) | 2.374 (±0.683) | 1.099 (±0.324) | MICA | 2022-04-13 |
| Learning an Animatable Detailed 3D Face Model from In-The-Wild Images | ✓ Link | 2.210 | 2.138 (±0.461) | 2.802 (±0.868) | 2.457 (±0.559) | 1.443 (±0.498) | DECA-f | 2020-12-07 |
| Learning to Regress 3D Face Shape and Expression from an Image without 3D Supervision | ✓ Link | 2.258 | 1.934 (±0.458) | 2.074 (±0.616) | 2.995 (±0.908) | 2.028 (±0.720) | RingNet | 2019-05-16 |
| ExpNet: Landmark-Free, Deep, 3D Facial Expressions | ✓ Link | 2.306 | 2.509 (±0.486) | 1.912 (±0.450) | 3.084 (±1.005) | 1.717 (±0.590) | ExpNet | 2018-02-02 |
| EMOCA: Emotion Driven Monocular Face Capture and Animation | ✓ Link | 2.388 | 2.532 (±0.539) | 2.929 (±1.106) | 2.595 (±0.631) | 1.495 (±0.469) | EMOCA-f | 2022-04-24 |
| Towards High-fidelity Nonlinear 3D Face Morphable Model | | 2.953 | 2.936 (±0.810) | 2.375 (±0.599) | 4.582 (±1.488) | 1.918 (±0.801) | N-3DMM | 2019-04-09 |