SRFlow: A Dataset and Regularization Model for High-Resolution Facial Optical Flow via Splatting Rasterization

TL;DR

This paper introduces SRFlow, a high-resolution facial optical flow dataset, and SRFlowNet, a model with specialized regularization that significantly improves flow estimation accuracy and micro-expression recognition.

Contribution

The paper presents the first high-resolution facial optical flow dataset and a novel regularized model tailored for high-resolution skin motion estimation.

Findings

Training with SRFlow dataset reduces EPE by up to 42%.

SRFlowNet achieves up to 48% improvement in F1-score on micro-expression datasets.

Regularization effectively suppresses noise in texture-less regions.

Abstract

Facial optical flow supports a wide range of tasks in facial motion analysis. However, the lack of high-resolution facial optical flow datasets has hindered progress in this area. In this paper, we introduce Splatting Rasterization Flow (SRFlow), a high-resolution facial optical flow dataset, and Splatting Rasterization Guided FlowNet (SRFlowNet), a facial optical flow model with tailored regularization losses. These losses constrain flow predictions using masks and gradients computed via difference or Sobel operator. This effectively suppresses high-frequency noise and large-scale errors in texture-less or repetitive-pattern regions, enabling SRFlowNet to be the first model explicitly capable of capturing high-resolution skin motion guided by Gaussian splatting rasterization. Experiments show that training with the SRFlow dataset improves facial optical flow estimation across various optical flow models, reducing end-point error (EPE) by up to 42% (from 0.5081 to 0.2953). Furthermore, when coupled with the SRFlow dataset, SRFlowNet achieves up to a 48% improvement in F1-score (from 0.4733 to 0.6947) on a composite of three micro-expression datasets. These results demonstrate the value of advancing both facial optical flow estimation and micro-expression recognition.