Semantic Image Matting

TL;DR

This paper introduces a semantic-aware approach to image matting that incorporates semantic classification and a semantic trimap to improve alpha matte quality, supported by a new large-scale dataset.

Contribution

It proposes extending traditional trimaps to semantic trimaps, learning a multi-class discriminator, and creating a large-scale semantic image matting dataset, advancing the state-of-the-art.

Findings

Outperforms existing methods on multiple benchmarks.

Achieves state-of-the-art performance in image matting.

Provides a large, balanced semantic image matting dataset.

Abstract

Natural image matting separates the foreground from background in fractional occupancy which can be caused by highly transparent objects, complex foreground (e.g., net or tree), and/or objects containing very fine details (e.g., hairs). Although conventional matting formulation can be applied to all of the above cases, no previous work has attempted to reason the underlying causes of matting due to various foreground semantics. We show how to obtain better alpha mattes by incorporating into our framework semantic classification of matting regions. Specifically, we consider and learn 20 classes of matting patterns, and propose to extend the conventional trimap to semantic trimap. The proposed semantic trimap can be obtained automatically through patch structure analysis within trimap regions. Meanwhile, we learn a multi-class discriminator to regularize the alpha prediction at semantic level, and content-sensitive weights to balance different regularization losses. Experiments on multiple benchmarks show that our method outperforms other methods and has achieved the most competitive state-of-the-art performance. Finally, we contribute a large-scale Semantic Image Matting Dataset with careful consideration of data balancing across different semantic classes.