ExtremeC3Net: Extreme Lightweight Portrait Segmentation Networks using Advanced C3-modules

TL;DR

ExtremeC3Net introduces an ultra-lightweight portrait segmentation model with a novel architecture that significantly reduces parameters while maintaining high accuracy, and includes new dataset annotations to address bias and improve performance.

Contribution

The paper presents a highly efficient portrait segmentation network using advanced C3-modules, reducing parameters by over 98%, and introduces new dataset annotations to analyze bias and enhance accuracy.

Findings

Reduces model parameters from 2.1M to 37.7K with minimal accuracy loss.

Outperforms existing lightweight segmentation models on EG1800 dataset.

Provides new annotated data addressing bias in portrait datasets.

Abstract

Designing a lightweight and robust portrait segmentation algorithm is an important task for a wide range of face applications. However, the problem has been considered as a subset of the object segmentation problem. bviously, portrait segmentation has its unique requirements. First, because the portrait segmentation is performed in the middle of a whole process of many realworld applications, it requires extremely lightweight models. Second, there has not been any public datasets in this domain that contain a sufficient number of images with unbiased statistics. To solve the problems, we introduce a new extremely lightweight portrait segmentation model consisting of a two-branched architecture based on the concentrated-comprehensive convolutions block. Our method reduces the number of parameters from 2.1M to 37.7K (around 98.2% reduction), while maintaining the accuracy within a 1% margin from the state-of-the-art portrait segmentation method. In our qualitative and quantitative analysis on the EG1800 dataset, we show that our method outperforms various existing lightweight segmentation models. Second, we propose a simple method to create additional portrait segmentation data which can improve accuracy on the EG1800 dataset. Also, we analyze the bias in public datasets by additionally annotating race, gender, and age on our own. The augmented dataset, the additional annotations and code are available in https://github.com/HYOJINPARK/ExtPortraitSeg .