We begin with 8 input images from wide-baseline viewpoints, extract and refine the 3D strands (the strands from 3 adjacent views are shown in different colors), and reconstruct the surface from the positions of the refined strands. Our result is robust to the wide-baseline setup and reveals detailed hair structures. In contrast, general multi-view stereo methods based on texture are less accurate [5] or unable to converge in the wide-baseline setup [7].
We propose a novel algorithm to reconstruct the 3D geometry of human hairs in wide-baseline setups using strand-based refinement. The hair strands are first extracted in each 2D view, and projected onto the 3D visual hull for initialization. The 3D positions of these strands are then refined by optimizing an objective function that takes into account cross-view hair orientation consistency, the visual hull constraint and smoothness constraints defined at the strand, wisp and global levels. Based on the refined strands, the algorithm can reconstruct an approximate hair surface: experiments with synthetic hair models achieve an accuracy of ~3mm. We also show real-world examples to demonstrate the capability to capture full-head hair styles as well as hair in motion with as few as 8 cameras.