3D Face Reconstruction with the Geometric Guidance of Facial Part Segmentation

TL;DR

This paper introduces a novel geometric guidance method using facial part segmentation for 3D face reconstruction, achieving state-of-the-art results by employing the Part Re-projection Distance Loss (PRDL) to improve geometric accuracy.

Contribution

The paper proposes PRDL, a new loss function leveraging facial segmentation geometry, which enhances 3D face reconstruction accuracy over existing renderer-based methods.

Findings

PRDL provides clear gradients for optimization.

Achieves state-of-the-art reconstruction performance.

Demonstrates robustness on faces with extreme expressions.

Abstract

3D Morphable Models (3DMMs) provide promising 3D face reconstructions in various applications. However, existing methods struggle to reconstruct faces with extreme expressions due to deficiencies in supervisory signals, such as sparse or inaccurate landmarks. Segmentation information contains effective geometric contexts for face reconstruction. Certain attempts intuitively depend on differentiable renderers to compare the rendered silhouettes of reconstruction with segmentation, which is prone to issues like local optima and gradient instability. In this paper, we fully utilize the facial part segmentation geometry by introducing Part Re-projection Distance Loss (PRDL). Specifically, PRDL transforms facial part segmentation into 2D points and re-projects the reconstruction onto the image plane. Subsequently, by introducing grid anchors and computing different statistical distances from these anchors to the point sets, PRDL establishes geometry descriptors to optimize the distribution of the point sets for face reconstruction. PRDL exhibits a clear gradient compared to the renderer-based methods and presents state-of-the-art reconstruction performance in extensive quantitative and qualitative experiments. Our project is available at https://github.com/wang-zidu/3DDFA-V3 .