S2F2: Self-Supervised High Fidelity Face Reconstruction from Monocular Image
Not on the map. Why?
Abstract (arXiv)
We present a novel face reconstruction method capable of reconstructing detailed face geometry, spatially varying face reflectance from a single monocular image. We build our work upon the recent advances of DNN-based auto-encoders with differentiable ray tracing image formation, trained in self-supervised manner. While providing the advantage of learning-based approaches and real-time reconstruction, the latter methods lacked fidelity. In this work, we achieve, for the first time, high fidelity face reconstruction using self-supervised learning only. Our novel coarse-to-fine deep architecture allows us to solve the challenging problem of decoupling face reflectance from geometry using a single image, at high computational speed. Compared to state-of-the-art methods, our method achieves more visually appealing reconstruction.
Key facts. Cited by 17 papers in the map (data through 7 Oct 2026). Builds on 29 of its 64 references that are in the map. First posted on arXiv 15 Mar 2022.
Builds on
29 of 64 references.
References · 29
- 2014Deep Learning Face Attributes in the Wildcited by 3,679
- 2020Face2Face: Real-time Face Capture and Reenactment of RGB Videoscited by 631
- 2017How far are we from solving the 2D & 3D Face Alignment problem? (and a dataset of 230,000 3D facial landmarks)cited by 524
- 2019Accurate 3D Face Reconstruction with Weakly-Supervised Learning: From Single Image to Image Setcited by 481
- 20193D Morphable Face Models -- Past, Present and Futurecited by 269
- 2020Towards Fast, Accurate and Stable 3D Dense Face Alignmentcited by 219
- 2020FaceScape: a Large-scale High Quality 3D Face Dataset and Detailed Riggable 3D Face Predictioncited by 196
- 2019Learning to Regress 3D Face Shape and Expression from an Image without 3D Supervisioncited by 176
- 2017MoFA: Model-based Deep Convolutional Face Autoencoder for Unsupervised Monocular Reconstructioncited by 163
- 2018Nonlinear 3D Face Morphable Modelcited by 157
- 2017Morphable Face Models - An Open Frameworkcited by 138
- 2017SfSNet: Learning Shape, Reflectance and Illuminance of Faces in the Wildcited by 110
17 more
- 2017Self-supervised Multi-level Face Model Learning for Monocular Reconstruction at over 250 Hzcited by 91
- 2019Towards High-fidelity Nonlinear 3D Face Morphable Modelcited by 91
- 2020AvatarMe: Realistically Renderable 3D Facial Reconstruction "in-the-wild"cited by 91
- 2018FML: Face Model Learning from Videoscited by 78
- 2016Learning Detailed Face Reconstruction from a Single Imagecited by 71
- 2019Towards a complete 3D morphable model of the human headcited by 63
- 2017Unrestricted Facial Geometry Reconstruction Using Image-to-Image Translationcited by 56
- 2020A Morphable Face Albedo Modelcited by 54
- 2021Towards High Fidelity Monocular Face Reconstruction with Rich Reflectance using Self-supervised Learning and Ray Tracingcited by 48
- 2020Portrait Shadow Manipulationcited by 42
- 2019Photo-Realistic Facial Details Synthesis from Single Imagecited by 40
- 2021Practical Face Reconstruction via Differentiable Ray Tracingcited by 35
- 20173D Face Reconstruction with Geometry Details from a Single Imagecited by 27
- 2020Cross-modal Deep Face Normals with Deactivable Skip Connectionscited by 20
- 2021Efficient and Differentiable Shadow Computation for Inverse Problemscited by 11
- 20193D Face Reconstruction from A Single Image Assisted by 2D Face Images in the Wildcited by 7
- 2019Face Reflectance and Geometry Modeling via Differentiable Ray Tracingcited by 4
Built on by
Citing papers per year: 2022: 1; 2023: 7; 2024: 3; 2025: 3; 2026: 3.
17 papers.
Sort:
By area:
- 2022High-Res Facial Appearance Capture from Polarized Smartphone Imagescited by 21
- 2023Learning a 3D Morphable Face Reflectance Model from Low-cost Datacited by 20
- 2023MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shadingcited by 18
- 2023Text-Guided 3D Face Synthesis -- From Generation to Editingcited by 8
- 2023A vector quantized masked autoencoder for audiovisual speech emotion recognitioncited by 6
- 2023DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignmentcited by 6
- 2023JoIN: Joint GANs Inversion for Intrinsic Image Decompositioncited by 4
- 2024Monocular Facial Appearance Capture in the Wildcited by 2
- 2024Democratizing the Creation of Animatable Facial Avatarscited by 1
- 2024Learning to Decouple the Lights for 3D Face Texture Modelingcited by 1
- 2025AvatarTex: High-Fidelity Facial Texture Reconstruction from Single-Image Stylized Avatarscited by 1
- 2026Skullptor: High Fidelity 3D Head Reconstruction in Seconds with Multi-View Normal Predictioncited by 1
5 more
- 2023HeadRecon: High-Fidelity 3D Head Reconstruction from Monocular Videocited by 0
- 2025Improving Facial Rig Semantics for Tracking and Retargetingcited by 0
- 2025Using Gaussian Splats to Create High-Fidelity Facial Geometry and Texturecited by 0
- 2026RelightAnyone: A Generalized Relightable 3D Gaussian Head Modelcited by 0
- 2026FaceSnap: Real-Time Personalized Lightstage Facial Performance Capturecited by 0
- 2026FaceSnap: Real-Time Personalized Lightstage Facial Performance Capture31 Aug 2026
- 2026Skullptor: High Fidelity 3D Head Reconstruction in Seconds with Multi-View Normal Prediction24 Feb 2026
- 2026RelightAnyone: A Generalized Relightable 3D Gaussian Head Model6 Jan 2026
- 2025Using Gaussian Splats to Create High-Fidelity Facial Geometry and Texture18 Dec 2025
- 2025AvatarTex: High-Fidelity Facial Texture Reconstruction from Single-Image Stylized Avatars10 Nov 2025
- 2025Improving Facial Rig Semantics for Tracking and Retargeting11 Aug 2025
- 2024Monocular Facial Appearance Capture in the Wild17 Dec 2024
- 2024Learning to Decouple the Lights for 3D Face Texture Modeling11 Dec 2024
- 2024Democratizing the Creation of Animatable Facial Avatars29 Jan 2024
- 2023MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shading20 Dec 2023
- 2023HeadRecon: High-Fidelity 3D Head Reconstruction from Monocular Video14 Dec 2023
- 2023Text-Guided 3D Face Synthesis -- From Generation to Editing1 Dec 2023
5 more
- 2023DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignment19 May 2023
- 2023JoIN: Joint GANs Inversion for Intrinsic Image Decomposition18 May 2023
- 2023A vector quantized masked autoencoder for audiovisual speech emotion recognition5 May 2023
- 2023Learning a 3D Morphable Face Reflectance Model from Low-cost Data21 Mar 2023
- 2022High-Res Facial Appearance Capture from Polarized Smartphone Images2 Dec 2022
- 2022High-Res Facial Appearance Capture from Polarized Smartphone Imagescited by 21
- 2023Learning a 3D Morphable Face Reflectance Model from Low-cost Datacited by 20
- 2023MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shadingcited by 18
- 2023A vector quantized masked autoencoder for audiovisual speech emotion recognitioncited by 6
- 2023DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignmentcited by 6
- 2023JoIN: Joint GANs Inversion for Intrinsic Image Decompositioncited by 4
- 2024Monocular Facial Appearance Capture in the Wildcited by 2
- 2024Democratizing the Creation of Animatable Facial Avatarscited by 1
- 2024Learning to Decouple the Lights for 3D Face Texture Modelingcited by 1
- 2025AvatarTex: High-Fidelity Facial Texture Reconstruction from Single-Image Stylized Avatarscited by 1
- 2026Skullptor: High Fidelity 3D Head Reconstruction in Seconds with Multi-View Normal Predictioncited by 1
- 2023HeadRecon: High-Fidelity 3D Head Reconstruction from Monocular Videocited by 0
12 most cited of 16 All 16 (JSON)
- 2026FaceSnap: Real-Time Personalized Lightstage Facial Performance Capture31 Aug 2026
- 2026Skullptor: High Fidelity 3D Head Reconstruction in Seconds with Multi-View Normal Prediction24 Feb 2026
- 2026RelightAnyone: A Generalized Relightable 3D Gaussian Head Model6 Jan 2026
- 2025Using Gaussian Splats to Create High-Fidelity Facial Geometry and Texture18 Dec 2025
- 2025AvatarTex: High-Fidelity Facial Texture Reconstruction from Single-Image Stylized Avatars10 Nov 2025
- 2025Improving Facial Rig Semantics for Tracking and Retargeting11 Aug 2025
- 2024Monocular Facial Appearance Capture in the Wild17 Dec 2024
- 2024Learning to Decouple the Lights for 3D Face Texture Modeling11 Dec 2024
- 2024Democratizing the Creation of Animatable Facial Avatars29 Jan 2024
- 2023MoSAR: Monocular Semi-Supervised Model for Avatar Reconstruction using Differentiable Shading20 Dec 2023
- 2023HeadRecon: High-Fidelity 3D Head Reconstruction from Monocular Video14 Dec 2023
- 2023DSFNet: Dual Space Fusion Network for Occlusion-Robust 3D Dense Face Alignment19 May 2023
12 newest of 16 All 16 (JSON)