Semantic Human Matting

TL;DR

This paper introduces SHM, an automatic human matting algorithm that leverages learned semantic constraints and deep networks to produce high-quality alpha mattes without user input, enabling scalable and efficient human extraction.

Contribution

It is the first to jointly learn semantic information and fine details for human matting using deep networks, removing the need for user-provided trimaps or scribbles.

Findings

SHM achieves comparable results to state-of-the-art interactive methods.

A large dataset with 35,513 annotated foregrounds was created for training and evaluation.

Extensive experiments demonstrate SHM's effectiveness on real images.

Abstract

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.