C-DLinkNet: considering multi-level semantic features for human parsing

TL;DR

This paper introduces C-DLinkNet, an end-to-end human parsing model that effectively combines multi-level features to improve semantic segmentation accuracy without extra data or large inputs.

Contribution

The paper presents C-DLinkNet with a novel Smooth Module for multi-level feature fusion, achieving competitive results with smaller inputs and no additional information.

Findings

Achieves mIoU=53.05 on LIP dataset validation set.

Outperforms some state-of-the-art methods with smaller input sizes.

Uses a new Smooth Module for feature combination.

Abstract

Human parsing is an essential branch of semantic segmentation, which is a fine-grained semantic segmentation task to identify the constituent parts of human. The challenge of human parsing is to extract effective semantic features to resolve deformation and multi-scale variations. In this work, we proposed an end-to-end model called C-DLinkNet based on LinkNet, which contains a new module named Smooth Module to combine the multi-level features in Decoder part. C-DLinkNet is capable of producing competitive parsing performance compared with the state-of-the-art methods with smaller input sizes and no additional information, i.e., achiving mIoU=53.05 on the validation set of LIP dataset.