# FFAvatar: Few-Shot, Feed-Forward, and Generalizable Avatar Reconstruction

arXiv: https://arxiv.org/abs/2605.15320 · HTML: https://mlsanity.com/paper/2605.15320/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 33 references that are in the map.
- First posted on arXiv 14 May 2026.

## Builds on (most cited first, 26 of 26)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Models](https://mlsanity.com/paper/2301.12597/) (2023)
- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [LRM: Large Reconstruction Model for Single Image to 3D](https://mlsanity.com/paper/2311.04400/) (2023)
- [Face2Face: Real-time Face Capture and Reenactment of RGB Videos](https://mlsanity.com/paper/2007.14808/) (2020)
- [Learning an Animatable Detailed 3D Face Model from In-The-Wild Images](https://mlsanity.com/paper/2012.04012/) (2020)
- [Dynamic Neural Radiance Fields for Monocular 4D Facial Avatar Reconstruction](https://mlsanity.com/paper/2012.03065/) (2020)
- [Magic123: One Image to High-Quality 3D Object Generation Using Both 2D and 3D Diffusion Priors](https://mlsanity.com/paper/2306.17843/) (2023)
- [Differentiable Augmentation for Data-Efficient GAN Training](https://mlsanity.com/paper/2006.10738/) (2020)
- [GaussianAvatars: Photorealistic Head Avatars with Rigged 3D Gaussians](https://mlsanity.com/paper/2312.02069/) (2023)
- [EMOCA: Emotion Driven Monocular Face Capture and Animation](https://mlsanity.com/paper/2204.11312/) (2022)
- [I M Avatar: Implicit Morphable Head Avatars from Videos](https://mlsanity.com/paper/2112.07471/) (2021)
- [Sapiens: Foundation for Human Vision Models](https://mlsanity.com/paper/2408.12569/) (2024)
- [HeadNeRF: A Real-time NeRF-based Parametric Head Model](https://mlsanity.com/paper/2112.05637/) (2021)
- [Projected GANs Converge Faster](https://mlsanity.com/paper/2111.01007/) (2021)
- [SplattingAvatar: Realistic Real-Time Human Avatars with Mesh-Embedded Gaussian Splatting](https://mlsanity.com/paper/2403.05087/) (2024)
- [Instant Volumetric Head Avatars](https://mlsanity.com/paper/2211.12499/) (2022)
- [Realistic One-shot Mesh-based Head Avatars](https://mlsanity.com/paper/2206.08343/) (2022)
- [Generalizable and Animatable Gaussian Head Avatar](https://mlsanity.com/paper/2410.07971/) (2024)
- [GPAvatar: Generalizable and Precise Head Avatar from Image(s)](https://mlsanity.com/paper/2401.10215/) (2024)
- [Avat3r: Large Animatable Gaussian Reconstruction Model for High-fidelity 3D Head Avatars](https://mlsanity.com/paper/2502.20220/) (2025)
- [LAM: Large Avatar Model for One-shot Animatable Gaussian Head](https://mlsanity.com/paper/2502.17796/) (2025)
- [Omni-ID: Holistic Identity Representation Designed for Generative Tasks](https://mlsanity.com/paper/2412.09694/) (2024)
- [ComposeMe: Attribute-Specific Image Prompts for Controllable Human Image Generation](https://mlsanity.com/paper/2509.18092/) (2025)

## Built on by (most cited first, 1 of 1)

- [RegHead: Non-Humanoid Head Blendshapes via Feed-Forward Registration](https://mlsanity.com/paper/2607.12206/) (2026)

## Abstract (arXiv)

Avatar reconstruction has traditionally relied on per-subject optimization that requires hours of computation or on expensive preprocessing that limits scalability. We introduce FFAvatar, a generalizable feed-forward framework that reconstructs high-quality, animatable 3D Gaussian head avatars from few-shot unposed portrait images in seconds. FFAvatar fuses information from multiple source images into a unified canonical Gaussian representation through Multi-View Query-Former, which is animated via FLAME parameters predicted end-to-end directly from pixels, eliminating the overhead of offline FLAME extraction. We further propose a three-stage training curriculum that achieves both broad generalization and high-fidelity reconstruction: (i) scalable pretraining on extensive monocular video data with over 1M identities to learn strong generalizable priors; (ii) multi-view fine-tuning on a small but high-quality dataset of 360-degree captures to enhance geometric fidelity and extreme-view awareness; and (iii) optional personalization that adapts to specific identities for maximum fidelity within 500 optimization steps. Extensive experiments demonstrate that FFAvatar sets a new standard for identity preservation, geometric consistency, and animation fidelity. On the NeRSemble benchmark, it outperforms the state-of-the-art LAM by a substantial 5.5 PSNR gain. Furthermore, FFAvatar enables real-time deployment, reconstructing avatars in 2 seconds without personalization and 10 seconds with personalization, while supporting 49 FPS animation on a single NVIDIA A100 GPU.