Paper Detail

PixelSmile: Toward Fine-Grained Facial Expression Editing

Jiabin Hua, Hengyuan Xu, Aojie Li, Wei Cheng, Gang Yu, Xingjun Ma, Yu-Gang Jiang

arxiv Score 9.6

Published 2026-03-26 · First seen 2026-03-27

General AI

Abstract

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.

Workflow Status

Review status
pending
Role
unreviewed
Read priority
soon
Vote
Not set.
Saved
no
Collections
Not filed yet.
Next action
Not filled yet.

Reading Brief

No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.

Why It Surfaced

No ranking explanation is available yet.

Tags

No tags.

BibTeX

@article{hua2026pixelsmile,
  title = {PixelSmile: Toward Fine-Grained Facial Expression Editing},
  author = {Jiabin Hua and Hengyuan Xu and Aojie Li and Wei Cheng and Gang Yu and Xingjun Ma and Yu-Gang Jiang},
  year = {2026},
  abstract = {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. PixelSm},
  url = {https://arxiv.org/abs/2603.25728},
  keywords = {cs.CV, cs.AI, diffusion framework, facial expression editing, semantic overlap, FFE dataset, FFE-Bench, PixelSmile, joint training, contrastive learning, expression disentanglement, textual latent interpolation, expression blending, code available, huggingface daily},
  eprint = {2603.25728},
  archiveprefix = {arXiv},
}

Metadata

{}