# Deformable Model-Driven Neural Rendering for High-Fidelity 3D Reconstruction of Human Heads Under Low-View Settings

arXiv: https://arxiv.org/abs/2303.13855 · HTML: https://mlsanity.com/paper/2303.13855/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 31 of its 52 references that are in the map.
- First posted on arXiv 24 Mar 2023.

## Builds on (most cited first, 30 of 31)

- [Adam: A Method for Stochastic Optimization](https://mlsanity.com/paper/1412.6980/) (2014)
- [PyTorch: An Imperative Style, High-Performance Deep Learning Library](https://mlsanity.com/paper/1912.01703/) (2019)
- [DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation](https://mlsanity.com/paper/1901.05103/) (2019)
- [Mip-NeRF: A Multiscale Representation for Anti-Aliasing Neural Radiance Fields](https://mlsanity.com/paper/2103.13415/) (2021)
- [NeuS: Learning Neural Implicit Surfaces by Volume Rendering for Multi-view Reconstruction](https://mlsanity.com/paper/2106.10689/) (2021)
- [Plenoxels: Radiance Fields without Neural Networks](https://mlsanity.com/paper/2112.05131/) (2021)
- [Volume Rendering of Neural Implicit Surfaces](https://mlsanity.com/paper/2106.12052/) (2021)
- [PlenOctrees for Real-time Rendering of Neural Radiance Fields](https://mlsanity.com/paper/2103.14024/) (2021)
- [Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations](https://mlsanity.com/paper/1906.01618/) (2019)
- [NeRF++: Analyzing and Improving Neural Radiance Fields](https://mlsanity.com/paper/2010.07492/) (2020)
- [Implicit Geometric Regularization for Learning Shapes](https://mlsanity.com/paper/2002.10099/) (2020)
- [Differentiable Volumetric Rendering: Learning Implicit 3D Representations without 3D Supervision](https://mlsanity.com/paper/1912.07372/) (2019)
- [UNISURF: Unifying Neural Implicit Surfaces and Radiance Fields for Multi-View Reconstruction](https://mlsanity.com/paper/2104.10078/) (2021)
- [3D Morphable Face Models -- Past, Present and Future](https://mlsanity.com/paper/1909.01815/) (2019)
- [I M Avatar: Implicit Morphable Head Avatars from Videos](https://mlsanity.com/paper/2112.07471/) (2021)
- [HeadNeRF: A Real-time NeRF-based Parametric Head Model](https://mlsanity.com/paper/2112.05637/) (2021)
- [Editing Conditional Radiance Fields](https://mlsanity.com/paper/2105.06466/) (2021)
- [Geo-Neus: Geometry-Consistent Neural Implicit Surfaces Learning for Multi-view Reconstruction](https://mlsanity.com/paper/2205.15848/) (2022)
- [SparseNeuS: Fast Generalizable Neural Surface Reconstruction from Sparse Views](https://mlsanity.com/paper/2206.05737/) (2022)
- [Improving neural implicit surfaces geometry with patch warping](https://mlsanity.com/paper/2112.09648/) (2021)
- [HF-NeuS: Improved Surface Reconstruction Using High-Frequency Details](https://mlsanity.com/paper/2206.07850/) (2022)
- [Deformed Implicit Field: Modeling 3D Shapes with Learned Dense Correspondence](https://mlsanity.com/paper/2011.13650/) (2020)
- [i3DMM: Deep Implicit 3D Morphable Model of Human Heads](https://mlsanity.com/paper/2011.14143/) (2020)
- [FaceVerse: a Fine-grained and Detail-controllable 3D Face Morphable Model from a Hybrid Dataset](https://mlsanity.com/paper/2203.14057/) (2022)
- [MoFaNeRF: Morphable Facial Neural Radiance Field](https://mlsanity.com/paper/2112.02308/) (2021)
- [H3D-Net: Few-Shot High-Fidelity 3D Head Reconstruction](https://mlsanity.com/paper/2107.12512/) (2021)
- [ImFace: A Nonlinear 3D Morphable Face Model with Implicit Neural Representations](https://mlsanity.com/paper/2203.14510/) (2022)
- [Geometry-Consistent Neural Shape Representation with Implicit Displacement Fields](https://mlsanity.com/paper/2106.05187/) (2021)
- [D2IM-Net: Learning Detail Disentangled Implicit Fields from Single Images](https://mlsanity.com/paper/2012.06650/) (2020)
- [Neural Strands: Learning Hair Geometry and Appearance from Multi-View Images](https://mlsanity.com/paper/2207.14067/) (2022)

## Built on by (most cited first, 7 of 7)

- [FatesGS: Fast and Accurate Sparse-View Surface Reconstruction using Gaussian Splatting with Depth-Feature Consistency](https://mlsanity.com/paper/2501.04628/) (2025)
- [SurFhead: Affine Rig Blending for Geometrically Accurate 2D Gaussian Surfel Head Avatars](https://mlsanity.com/paper/2410.11682/) (2024)
- [CoopTrack: Exploring End-to-End Learning for Efficient Cooperative Sequential Perception](https://mlsanity.com/paper/2507.19239/) (2025)
- [GSurf: Learning Signed Distance Fields from Splatting Opaque Gaussians for High-quality 3D Reconstruction](https://mlsanity.com/paper/2411.15723/) (2024)
- [Parameterization-driven Neural Surface Reconstruction for Object-oriented Editing in Neural Rendering](https://mlsanity.com/paper/2310.05524/) (2023)
- [SFDM: Robust Decomposition of Geometry and Reflectance for Realistic Face Rendering from Sparse-view Images](https://mlsanity.com/paper/2312.06085/) (2023)
- [Single Image, Any Face: Generalisable 3D Face Generation](https://mlsanity.com/paper/2409.16990/) (2024)

## Abstract (arXiv)

Reconstructing 3D human heads in low-view settings presents technical challenges, mainly due to the pronounced risk of overfitting with limited views and high-frequency signals. To address this, we propose geometry decomposition and adopt a two-stage, coarse-to-fine training strategy, allowing for progressively capturing high-frequency geometric details. We represent 3D human heads using the zero level-set of a combined signed distance field, comprising a smooth template, a non-rigid deformation, and a high-frequency displacement field. The template captures features that are independent of both identity and expression and is co-trained with the deformation network across multiple individuals with sparse and randomly selected views. The displacement field, capturing individual-specific details, undergoes separate training for each person. Our network training does not require 3D supervision or object masks. Experimental results demonstrate the effectiveness and robustness of our geometry decomposition and two-stage training strategy. Our method outperforms existing neural rendering approaches in terms of reconstruction accuracy and novel view synthesis under low-view settings. Moreover, the pre-trained template serves a good initialization for our model when encountering unseen individuals.