Colored Transparent Object Matting from a Single Image Using Deep Learning

TL;DR

This paper introduces a deep learning method for transparent object matting that handles colored transparent objects from a single image, extending previous work to improve speed and applicability in real-world scenarios.

Contribution

It extends the TOM-Net framework to model colored transparent objects using a color filter, enabling single-image matting of colored transparent objects with a new deep learning approach.

Findings

Effective on synthetic and real datasets

Produces promising matting results for colored transparent objects

Runs in a single fast feed-forward pass

Abstract

This paper proposes a deep learning based method for colored transparent object matting from a single image. Existing approaches for transparent object matting often require multiple images and long processing times, which greatly hinder their applications on real-world transparent objects. The recently proposed TOM-Net can produce a matte for a colorless transparent object from a single image in a single fast feed-forward pass. In this paper, we extend TOM-Net to handle colored transparent object by modeling the intrinsic color of a transparent object with a color filter. We formulate the problem of colored transparent object matting as simultaneously estimating an object mask, a color filter, and a refractive flow field from a single image, and present a deep learning framework for learning this task. We create a large-scale synthetic dataset for training our network. We also capture a real dataset for evaluation. Experiments on both synthetic and real datasets show promising results, which demonstrate the effectiveness of our method.