Fast Portrait Segmentation with Highly Light-weight Network

TL;DR

This paper introduces a fast, lightweight portrait segmentation method using a novel backbone architecture with fewer parameters, achieving high speed and competitive accuracy.

Contribution

The paper proposes a highly light-weight backbone with a bottleneck-based factorized block for efficient portrait segmentation.

Findings

Faster processing speed than existing methods

Maintains competitive accuracy on benchmark datasets

Uses fewer parameters for efficiency

Abstract

In this paper, we describe a fast and light-weight portrait segmentation method based on a new highly light-weight backbone (HLB) architecture. The core element of HLB is a bottleneck-based factorized block (BFB) that has much fewer parameters than existing alternatives while keeping good learning capacity. Consequently, the HLB-based portrait segmentation method can run faster than the existing methods yet retaining the competitive accuracy performance with state-of-the-arts. Experiments conducted on two benchmark datasets demonstrate the effectiveness and efficiency of our method.