# Cafca: High-quality Novel View Synthesis of Expressive Faces from Casual Few-shot Captures

arXiv: https://arxiv.org/abs/2410.00630 · HTML: https://mlsanity.com/paper/2410.00630/

## Key facts

- Cited by 34 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 19 references that are in the map.
- First posted on arXiv 1 Oct 2024.

## Builds on (most cited first, 10 of 10)

- [Mixture of Volumetric Primitives for Efficient Neural Rendering](https://mlsanity.com/paper/2103.01954/) (2021)
- [HyperNeRF: A Higher-Dimensional Representation for Topologically Varying Neural Radiance Fields](https://mlsanity.com/paper/2106.13228/) (2021)
- [NeRSemble: Multi-view Radiance Field Reconstruction of Human Heads](https://mlsanity.com/paper/2305.03027/) (2023)
- [Learning to Predict Indoor Illumination from a Single Image](https://mlsanity.com/paper/1704.00090/) (2017)
- [Real-Time Radiance Fields for Single-Image Portrait View Synthesis](https://mlsanity.com/paper/2305.02310/) (2023)
- [LatentAvatar: Learning Latent Expression Code for Expressive Neural Head Avatar](https://mlsanity.com/paper/2305.01190/) (2023)
- [FDNeRF: Few-shot Dynamic Neural Radiance Fields for Face Reconstruction and Expression Editing](https://mlsanity.com/paper/2208.05751/) (2022)
- [VoLux-GAN: A Generative Model for 3D Face Synthesis with HDRI Relighting](https://mlsanity.com/paper/2201.04873/) (2022)
- [HAvatar: High-fidelity Head Avatar via Facial Model Conditioned Neural Radiance Field](https://mlsanity.com/paper/2309.17128/) (2023)
- [BakedAvatar: Baking Neural Fields for Real-Time Head Avatar Synthesis](https://mlsanity.com/paper/2311.05521/) (2023)

## Built on by (most cited first, 30 of 34)

- [Avat3r: Large Animatable Gaussian Reconstruction Model for High-fidelity 3D Head Avatars](https://mlsanity.com/paper/2502.20220/) (2025)
- [Vid2Avatar-Pro: Authentic Avatar from Videos in the Wild via Universal Prior](https://mlsanity.com/paper/2503.01610/) (2025)
- [FaceLift: Learning Generalizable Single Image 3D Face Reconstruction from Synthetic Heads](https://mlsanity.com/paper/2412.17812/) (2024)
- [Synthetic Prior for Few-Shot Drivable Head Avatar Inversion](https://mlsanity.com/paper/2501.06903/) (2025)
- [FlexAvatar: Learning Complete 3D Head Avatars with Partial Supervision](https://mlsanity.com/paper/2512.15599/) (2025)
- [GASP: Gaussian Avatars with Synthetic Priors](https://mlsanity.com/paper/2412.07739/) (2024)
- [FastAvatar: Instant 3D Gaussian Splatting for Faces from Single Unconstrained Poses](https://mlsanity.com/paper/2508.18389/) (2025)
- [PercHead: Perceptual Head Model for Single-Image 3D Head Reconstruction & Editing](https://mlsanity.com/paper/2511.02777/) (2025)
- [Audio Driven Real-Time Facial Animation for Social Telepresence](https://mlsanity.com/paper/2510.01176/) (2025)
- [UIKA: Fast Universal Head Avatar from Pose-Free Images](https://mlsanity.com/paper/2601.07603/) (2026)
- [Dream, Lift, Animate: From Single Images to Animatable Gaussian Avatars](https://mlsanity.com/paper/2507.15979/) (2025)
- [Controlling Avatar Diffusion with Learnable Gaussian Embedding](https://mlsanity.com/paper/2503.15809/) (2025)
- [Parametric Gaussian Human Model: Generalizable Prior for Efficient and Realistic Human Avatar Modeling](https://mlsanity.com/paper/2506.06645/) (2025)
- [HyperGaussians: High-Dimensional Gaussian Splatting for High-Fidelity Animatable Face Avatars](https://mlsanity.com/paper/2507.02803/) (2025)
- [VoluMe -- Authentic 3D Video Calls from Live Gaussian Splat Prediction](https://mlsanity.com/paper/2507.21311/) (2025)
- [FlexAvatar: Flexible Large Reconstruction Model for Animatable Gaussian Head Avatars with Detailed Deformation](https://mlsanity.com/paper/2512.17717/) (2025)
- [TexAvatars : Hybrid Texel-3D Representations for Stable Rigging of Photorealistic Gaussian Head Avatars](https://mlsanity.com/paper/2512.21099/) (2025)
- [GNM Head: A Generative aNthropometric Model of the human head](https://mlsanity.com/paper/2607.23687/) (2026)
- [Pix2NPHM: Learning to Regress NPHM Reconstructions From a Single Image](https://mlsanity.com/paper/2512.17773/) (2025)
- [ELITE: Efficient Gaussian Head Avatar from a Monocular Video via Learned Initialization and TEst-time Generative Adaptation](https://mlsanity.com/paper/2601.10200/) (2026)
- [LegacyAvatars: Volumetric Face Avatars For Traditional Graphics Pipelines](https://mlsanity.com/paper/2601.12285/) (2026)
- [AvatarPointillist: AutoRegressive 4D Gaussian Avatarization](https://mlsanity.com/paper/2604.04787/) (2026)
- [Large-Scale High-Quality 3D Gaussian Head Reconstruction from Multi-View Captures](https://mlsanity.com/paper/2605.04035/) (2026)
- [Multi-view Consistent 3D Gaussian Head Avatars 'without' Multi-view Generation](https://mlsanity.com/paper/2605.25220/) (2026)
- [FiCA: Feed-forward instant Gaussian Codec Avatars from a Single Portrait Image](https://mlsanity.com/paper/2606.24232/) (2026)
- [High-Quality 3D Head Reconstruction from Any Single Portrait Image](https://mlsanity.com/paper/2503.08516/) (2025)
- [Low-Rank Head Avatar Personalization with Registers](https://mlsanity.com/paper/2506.01935/) (2025)
- [Improving Facial Rig Semantics for Tracking and Retargeting](https://mlsanity.com/paper/2508.08429/) (2025)
- [Using Gaussian Splats to Create High-Fidelity Facial Geometry and Texture](https://mlsanity.com/paper/2512.16397/) (2025)
- [3DRealHead: Few-Shot Detailed Head Avatar](https://mlsanity.com/paper/2604.13171/) (2026)

## Abstract (arXiv)

Volumetric modeling and neural radiance field representations have revolutionized 3D face capture and photorealistic novel view synthesis. However, these methods often require hundreds of multi-view input images and are thus inapplicable to cases with less than a handful of inputs. We present a novel volumetric prior on human faces that allows for high-fidelity expressive face modeling from as few as three input views captured in the wild. Our key insight is that an implicit prior trained on synthetic data alone can generalize to extremely challenging real-world identities and expressions and render novel views with fine idiosyncratic details like wrinkles and eyelashes. We leverage a 3D Morphable Face Model to synthesize a large training set, rendering each identity with different expressions, hair, clothing, and other assets. We then train a conditional Neural Radiance Field prior on this synthetic dataset and, at inference time, fine-tune the model on a very sparse set of real images of a single subject. On average, the fine-tuning requires only three inputs to cross the synthetic-to-real domain gap. The resulting personalized 3D model reconstructs strong idiosyncratic facial expressions and outperforms the state-of-the-art in high-quality novel view synthesis of faces from sparse inputs in terms of perceptual and photo-metric quality.