Feature-preserving detailed 3D face reconstruction from a single image

Li, Yue; Ma, Liqian; Fan, Haoqiang; Mitchell, Kenny

Abstract

Dense 3D face reconstruction plays a fundamental role in visual media production involving digital actors. We improve upon high fidelity reconstruction from a single 2D photo with a reconstruction framework that is robust to large variations in expressions, poses and illumination. We provide a global optimization step improving the alignment of 3D facial geometry to tracked 2D landmarks with 3D Laplacian deformation. Face detail is improved through, extending Shape from Shading reconstruction with fitted albedo prior masks, together with a fast proportionality constraint between depth and image gradients consistent with local self-occlusion behavior. Together these measures better preserve the crucial facial features that define an actor's identity, and we illustrate this through a variety of comparisons with related works.

Citation

Li, Y., Ma, L., Fan, H., & Mitchell, K. (2018, December). Feature-preserving detailed 3D face reconstruction from a single image. Presented at the 15th ACM SIGGRAPH European Conference, London, United Kingdom

• Presentation Conference Type: Conference Paper (published)
• Conference Name: the 15th ACM SIGGRAPH European Conference
• Start Date: Dec 13, 2018
• End Date: Dec 14, 2018
• Acceptance Date: Sep 17, 2018
• Publication Date: Dec 13, 2018
• Deposit Date: Dec 4, 2018
• Publicly Available Date: Dec 5, 2018
• Publisher: Association for Computing Machinery (ACM)
• Series Title: ACM International Conference Proceeding Series
• Book Title: CVMP '18 Proceedings of the 15th ACM SIGGRAPH European Conference on Visual Media Production
• ISBN: 9781450360586
• DOI: https://doi.org/10.1145/3278471.3278473
• Keywords: Feature-Preserving, 3D Face Reconstruction, Optimization
• Public URL: http://researchrepository.napier.ac.uk/Output/1413576
• Contract Date: Dec 4, 2018