SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Kang, Zejian, Xu, Xuanyang, Yang, Wentao, Zheng, Kai, Fei, Yuanchen, Zou, Hongyuan, Shan, Hui, Yang, Shuo, Huang, Xiangru
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913098769629184
author Kang, Zejian
Xu, Xuanyang
Yang, Wentao
Zheng, Kai
Fei, Yuanchen
Zou, Hongyuan
Shan, Hui
Yang, Shuo
Huang, Xiangru
author_facet Kang, Zejian
Xu, Xuanyang
Yang, Wentao
Zheng, Kai
Fei, Yuanchen
Zou, Hongyuan
Shan, Hui
Yang, Shuo
Huang, Xiangru
contents Accurate facial estimation is crucial for realistic digital human animation, and ARKit blendshape coefficients offer an interpretable representation by mapping facial motions to semantic animation controls. However, learning high-quality ARKit coefficient prediction remains limited by the absence of reliable ground-truth supervision. Existing methods typically rely on capture software such as Live Link Face to provide pseudo labels, which may contain noisy activations, biased coefficient magnitudes, and missing or inaccurate facial actions. Consequently, models trained with supervised learning tend to reproduce imperfect pseudo labels rather than optimize for perceptual expression fidelity. In this paper, we propose SuperFace, a preference-driven framework that moves ARKit facial expression estimation from pseudo-label imitation toward human-aligned perceptual optimization. Instead of treating software-estimated coefficients as fixed ground truth, SuperFace uses them only as an initialization and further improves coefficient prediction through human preference feedback on rendered facial expressions. By aligning the model with perceptual judgments rather than numerical pseudo labels, SuperFace enables more visually faithful and expressive facial animation. Experiments show that SuperFace improves expression fidelity over Live Link Face supervision, demonstrating the effectiveness of preference-driven optimization for semantic facial action prediction.
format Preprint
id arxiv_https___arxiv_org_abs_2605_06179
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision
Kang, Zejian
Xu, Xuanyang
Yang, Wentao
Zheng, Kai
Fei, Yuanchen
Zou, Hongyuan
Shan, Hui
Yang, Shuo
Huang, Xiangru
Computer Vision and Pattern Recognition
Accurate facial estimation is crucial for realistic digital human animation, and ARKit blendshape coefficients offer an interpretable representation by mapping facial motions to semantic animation controls. However, learning high-quality ARKit coefficient prediction remains limited by the absence of reliable ground-truth supervision. Existing methods typically rely on capture software such as Live Link Face to provide pseudo labels, which may contain noisy activations, biased coefficient magnitudes, and missing or inaccurate facial actions. Consequently, models trained with supervised learning tend to reproduce imperfect pseudo labels rather than optimize for perceptual expression fidelity. In this paper, we propose SuperFace, a preference-driven framework that moves ARKit facial expression estimation from pseudo-label imitation toward human-aligned perceptual optimization. Instead of treating software-estimated coefficients as fixed ground truth, SuperFace uses them only as an initialization and further improves coefficient prediction through human preference feedback on rendered facial expressions. By aligning the model with perceptual judgments rather than numerical pseudo labels, SuperFace enables more visually faithful and expressive facial animation. Experiments show that SuperFace improves expression fidelity over Live Link Face supervision, demonstrating the effectiveness of preference-driven optimization for semantic facial action prediction.
title SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.06179