LEED: Label-Free Expression Editing via Disentanglement

TL;DR

LEED introduces a label-free facial expression editing framework that disentangles identity and expression without requiring expression labels, enabling high-quality editing of frontal and profile faces.

Contribution

The paper proposes a novel disentanglement-based method for expression editing that operates without labeled data, improving flexibility and reducing annotation costs.

Findings

LEED achieves superior qualitative expression editing results.

LEED outperforms existing methods quantitatively on public datasets.

The framework effectively disentangles identity and expression without labels.

Abstract

Recent studies on facial expression editing have obtained very promising progress. On the other hand, existing methods face the constraint of requiring a large amount of expression labels which are often expensive and time-consuming to collect. This paper presents an innovative label-free expression editing via disentanglement (LEED) framework that is capable of editing the expression of both frontal and profile facial images without requiring any expression label. The idea is to disentangle the identity and expression of a facial image in the expression manifold, where the neutral face captures the identity attribute and the displacement between the neutral image and the expressive image captures the expression attribute. Two novel losses are designed for optimal expression disentanglement and consistent synthesis, including a mutual expression information loss that aims to extract pure expression-related features and a siamese loss that aims to enhance the expression similarity between the synthesized image and the reference image. Extensive experiments over two public facial expression datasets show that LEED achieves superior facial expression editing qualitatively and quantitatively.