no code implementations • 7 Mar 2019 • Yuanxi Ma, Cen Wang, Shiying Li, Jingyi Yu
Robust segmentation of hair from portrait images remains challenging: hair does not conform to a uniform shape, style or even color; dark hair in particular lacks features.
no code implementations • 26 Mar 2018 • Huangjie Yu, Guli Zhang, Yuanxi Ma, Yingliang Zhang, Jingyi Yu
We present a novel semantic light field (LF) refocusing technique that can achieve unprecedented see-through quality.
no code implementations • ICCV 2017 • Yingliang Zhang, Peihong Yu, Wei Yang, Yuanxi Ma, Jingyi Yu
In this paper, we explore using light fields captured by plenoptic cameras or camera arrays as inputs.