PixelSmile: Toward Fine-Grained Facial Expression Editing

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hua, Jiabin, Xu, Hengyuan, Li, Aojie, Cheng, Wei, Yu, Gang, Ma, Xingjun, Jiang, Yu-Gang
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914425630359552
author Hua, Jiabin
Xu, Hengyuan
Li, Aojie
Cheng, Wei
Yu, Gang
Ma, Xingjun
Jiang, Yu-Gang
author_facet Hua, Jiabin
Xu, Hengyuan
Li, Aojie
Cheng, Wei
Yu, Gang
Ma, Xingjun
Jiang, Yu-Gang
contents 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 learning 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 identity preservation, confirming its effectiveness for continuous, controllable, and fine-grained expression editing, while naturally supporting smooth expression blending.
format Preprint
id arxiv_https___arxiv_org_abs_2603_25728
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle PixelSmile: Toward Fine-Grained Facial Expression Editing
Hua, Jiabin
Xu, Hengyuan
Li, Aojie
Cheng, Wei
Yu, Gang
Ma, Xingjun
Jiang, Yu-Gang
Computer Vision and Pattern Recognition
Artificial Intelligence
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 learning 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 identity preservation, confirming its effectiveness for continuous, controllable, and fine-grained expression editing, while naturally supporting smooth expression blending.
title PixelSmile: Toward Fine-Grained Facial Expression Editing
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2603.25728