PixelSmile: Toward Fine-Grained Facial Expression Editing
This work addresses the challenge of continuous and controllable facial expression editing for applications in computer vision and graphics, representing an incremental improvement over existing methods.
The paper tackles the problem of fine-grained facial expression editing by addressing semantic overlap, resulting in a method that achieves superior disentanglement and robust identity preservation with precise linear control.
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.