Lightweight Alpha Matting Network Using Distillation-Based Channel Pruning

TL;DR

This paper introduces a distillation-based channel pruning method to create lightweight alpha matting networks suitable for mobile devices, demonstrating improved performance and versatility across tasks.

Contribution

It proposes a novel channel pruning technique based on knowledge distillation, enhancing lightweight alpha matting models and applying the method to semantic segmentation.

Findings

Outperforms existing lightweight alpha matting methods

Reduces model complexity while maintaining accuracy

Proves versatility by applying to semantic segmentation

Abstract

Recently, alpha matting has received a lot of attention because of its usefulness in mobile applications such as selfies. Therefore, there has been a demand for a lightweight alpha matting model due to the limited computational resources of commercial portable devices. To this end, we suggest a distillation-based channel pruning method for the alpha matting networks. In the pruning step, we remove channels of a student network having fewer impacts on mimicking the knowledge of a teacher network. Then, the pruned lightweight student network is trained by the same distillation loss. A lightweight alpha matting model from the proposed method outperforms existing lightweight methods. To show superiority of our algorithm, we provide various quantitative and qualitative experiments with in-depth analyses. Furthermore, we demonstrate the versatility of the proposed distillation-based channel pruning method by applying it to semantic segmentation.