ExpressEdit: Fast Editing of Stylized Facial Expressions with Diffusion Models in Photoshop

2026-04-03

AI Summary

This work addresses the limitations of existing AI-based image editing methods, which often introduce global noise and pixel drift when stylizing facial expressions, thereby hindering integration into professional imaging workflows. To overcome these challenges, the authors propose a Photoshop plugin grounded in diffusion models, augmented with retrieval-enhanced generation and a curated database of 135 narrative-driven facial expression exemplars. The system enables high-fidelity, low-artifact expression editing while seamlessly interoperating with native tools such as Liquify. Leveraging GPU-accelerated inference, it achieves rapid editing—under three seconds on a single consumer-grade GPU—outperforming prevailing closed-source alternatives. Both the implementation code and the expression dataset are publicly released to foster reproducibility and further research.

Abstract

Facial expressions of characters are a vital component of visual storytelling. While current AI image editing models hold promise for assisting artists in the task of stylized expression editing, these models introduce global noise and pixel drift into the edited image, preventing the integration of these models into professional image editing software and workflows. To bridge this gap, we introduce ExpressEdit, a fully open-source Photoshop plugin that is free from common artifacts of proprietary image editing models and robustly synergizes with native Photoshop operations such as Liquify. ExpressEdit seamlessly edits an expression within 3 seconds on a single consumer-grade GPU, significantly faster than popular proprietary models. Moreover, to support the generation of diverse expressions according to different narrative needs, we compile a comprehensive expression database of 135 expression tags enriched with example stories and images designed for retrieval-augmented generation. We open source the code and dataset to facilitate future research and artistic exploration.

Problem

• facial expression editing
• image editing artifacts
• diffusion models
• professional image editing
• stylized expressions

Innovation

• diffusion models
• facial expression editing
• Photoshop plugin
• retrieval-augmented generation
• artifact-free editing