Semantic Human Matting

5 Sep 2018  ·  Quan Chen, Tiezheng Ge, Yanyu Xu, Zhiqiang Zhang, Xinxin Yang, Kun Gai ·

Human matting, high quality extraction of humans from natural images, is crucial for a wide variety of applications. Since the matting problem is severely under-constrained, most previous methods require user interactions to take user designated trimaps or scribbles as constraints. This user-in-the-loop nature makes them difficult to be applied to large scale data or time-sensitive scenarios. In this paper, instead of using explicit user input constraints, we employ implicit semantic constraints learned from data and propose an automatic human matting algorithm (SHM). SHM is the first algorithm that learns to jointly fit both semantic information and high quality details with deep networks. In practice, simultaneously learning both coarse semantics and fine details is challenging. We propose a novel fusion strategy which naturally gives a probabilistic estimation of the alpha matte. We also construct a very large dataset with high quality annotations consisting of 35,513 unique foregrounds to facilitate the learning and evaluation of human matting. Extensive experiments on this dataset and plenty of real images show that SHM achieves comparable results with state-of-the-art interactive matting methods.

PDF Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Image Matting AIM-500 SHM SAD 170.44 # 4
MSE 0.0921 # 5
MAD 0.1012 # 4
Conn. 170.67 # 4
Grad. 115.29 # 5
Image Matting AM-2K SHM SAD 17.81 # 5
MSE 0.0068 # 6
MAD 0.0102 # 5
Image Matting P3M-10k SHM SAD 21.56 # 3
MSE 0.0100 # 4
MAD 0.0125 # 3


No methods listed for this paper. Add relevant methods here