English(EN) SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision

SuperFace 利用人类偏好改进面部表情估计以用于动画制作

作者 PulseAugur 编辑部 · [2 个来源] · 2026-05-07 12:57

研究人员开发了 SuperFace，一个用于面部表情估计的新框架，它不再依赖不完美的软件生成标签。该系统使用人类偏好反馈来优化预测，旨在实现更具视觉真实感和表现力的数字动画。通过优先考虑感知判断而非伪标签，SuperFace 提高了 ARKit 混合形状系数预测的准确性。 AI

影响通过将面部表情模型与人类感知对齐，提高了数字人物动画的真实感。

排序理由这是一篇详细介绍面部表情估计新框架的研究论文。

AI 生成摘要 · Google Gemini · 来自 2 个来源。我们如何撰写摘要 →

报道来源 [2]

arXiv cs.CV TIER_1 English(EN) · Zejian Kang, Xuanyang Xu, Wentao Yang, Kai Zheng, Yuanchen Fei, Hongyuan Zou, Hui Shan, Shuo Yang, Xiangru Huang · 2026-05-08 04:00

SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision

arXiv:2605.06179v1 Announce Type: new Abstract: 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-qu…
arXiv cs.CV TIER_1 English(EN) · Xiangru Huang · 2026-05-07 12:57

SuperFace: Preference-Aligned Facial Expression Estimation Beyond Pseudo Supervision

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 limit…