Identity-Consistent Expression Fields: A Disentangled Neural Radiance Field Framework for Few-Shot Facial Expression Synthesis
Quick Answer
This paper shows that The Identity-Consistent Expression Fields (ICEF) framework enhances few-shot facial expression synthesis by disentangling identity-specific features from expression dynamics, improving identity preservation during novel expression rendering.
Quick Take
ICEF introduces a confidence-weighted warping mechanism to mitigate artifacts in expressions far from the training set, ensuring higher fidelity in facial animations.
Key Points
- ICEF disentangles static identity features from dynamic expression changes.
- Introduces a regularizer to preserve identity-specific appearance during expression edits.
- Confidence-weighted warping reduces artifacts in extrapolated expressions.
- Evaluates rendering quality and identity consistency across various expressions.
- Addresses limitations of prior few-shot dynamic NeRF methods.
DeepSignal Analysis
What happened
The Identity-Consistent Expression Fields (ICEF) framework addresses challenges in few-shot facial expression synthesis by separating identity features from expression dynamics. This approach enhances identity preservation during rendering of new expressions, utilizing a confidence-weighted warping mechanism to reduce artifacts in facial animations.
Key evidence
- ICEF disentangles a static identity-specific component from a dynamic expression-conditioned deformation component, which helps maintain identity during expression changes.
- The framework introduces an identity preservation regularizer that constrains modifications to expression-relevant areas, preserving the canonical appearance of the identity.
- A confidence-weighted warping step is included to down-weight unreliable warps for expressions that are significantly different from the training set, addressing issues seen in previous methods.
Why it matters
This framework is significant as it improves the quality of facial animations generated from limited input data, which is crucial for applications in virtual reality, gaming, and digital content creation. By ensuring that identity is preserved even when generating novel expressions, ICEF could enhance user experience and realism in interactive environments.
Paper Resources
📖 Reader Mode
~2 min readAbstract:Neural Radiance Fields (NeRF) have enabled photorealistic novel-view synthesis of 3D scenes and, in the facial domain, have been extended to reconstruct and animate 3D face models from a small number of images. However, existing few-shot dynamic NeRF methods for facial expression editing typically warp a single learned feature volume conditioned on target expression parameters, which can cause identity-specific appearance details (skin texture, fine geometric structure) to drift when the model is driven toward expressions far from those seen in the few-shot input set. We propose Identity-Consistent Expression Fields (ICEF), a framework that explicitly disentangles a static, identity-specific radiance component from a dynamic, expression-conditioned deformation component, and introduces an identity preservation regularizer that constrains the deformation network to modify only expression-relevant regions while leaving identity-specific canonical appearance untouched. ICEF further incorporates a confidence-weighted conditional feature warping step that down-weights unreliable warps for target expressions that are far, in parameter space, from the observed few-shot inputs, mitigating artifacts observed in prior few-shot dynamic NeRF methods when extrapolating to novel expressions. We relate ICEF to prior few-shot dynamic NeRF, static 3D-aware face generation, and disentangled face-editing radiance field methods, and describe an evaluation protocol measuring both novel-expression rendering quality and, specifically, identity-consistency metrics across a range of expression-parameter extrapolation distances.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.16287 [cs.CV] |
| (or arXiv:2607.16287v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2607.16287 arXiv-issued DOI via DataCite |
Submission history
From: Minh Tran Binh [view email]
[v1]
Sat, 11 Jul 2026 16:05:17 UTC (7 KB)
— Originally published at arxiv.org
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.