PixelSmile: Toward Fine-Grained Facial Expression Editing

论文概要

研究领域: 计算机视觉 作者: Jiabin Hua, Hengyuan Xu, Aojie Li, Wei Cheng, Gang Yu, Xingjun Ma, Yu-Gang Jiang 发布时间: 2026-03-26 arXiv: 2603.25728v1

中文摘要

细粒度面部表情编辑长期以来受到内在语义重叠的限制。为解决这一问题，我们构建了带有连续情感标注的Flex Facial Expression (FFE)数据集，并建立了FFE-Bench基准来评估结构混淆、编辑准确性、线性可控性以及表情编辑与身份保持之间的权衡。我们提出了PixelSmile，一个通过完全对称联合训练解耦表情语义的扩散框架。PixelSmile将强度监督与对比学习相结合，以产生更强、更可区分的表情，通过文本潜在插值实现精确稳定的线性表情控制。大量实验表明，PixelSmile实现了优越的解耦能力和鲁棒的身份保持，证实了其在连续、可控和细粒度表情编辑方面的有效性，同时自然支持平滑的表情混合。

原文摘要

Fine-grained facial expression editing has long been limited by intrinsic semantic overlap. To address this, we construct the Flex Facial Expression (FFE) dataset with continuous affective annotations and establish FFE-Bench to evaluate structural confusion, editing accuracy, linear controllability, and the trade-off between expression editing and identity preservation. We propose PixelSmile, a diffusion framework that disentangles expression semantics via fully symmetric joint training. PixelSmile combines intensity supervision with contrastive学习 to produce stronger and more distinguishable expressions, achieving precise and stable linear expression control through textual latent interpolation. Extensive experiments demonstrate that PixelSmile achieves superior disentanglement and robust ...

自动采集于 2026-03-28

#论文 #arXiv #计算机视觉 #小凯