Quality-Aware Network for Human Parsing

03/10/2021
by   Lu Yang, et al.
6

How to estimate the quality of the network output is an important issue, and currently there is no effective solution in the field of human parsing. In order to solve this problem, this work proposes a statistical method based on the output probability map to calculate the pixel quality information, which is called pixel score. In addition, the Quality-Aware Module (QAM) is proposed to fuse the different quality information, the purpose of which is to estimate the quality of human parsing results. We combine QAM with a concise and effective network design to propose Quality-Aware Network (QANet) for human parsing. Benefiting from the superiority of QAM and QANet, we achieve the best performance on three multiple and one single human parsing benchmarks, including CIHP, MHP-v2, Pascal-Person-Part and LIP. Without increasing the training and inference time, QAM improves the AP^r criterion by more than 10 points in the multiple human parsing task. QAM can be extended to other tasks with good quality estimation, e.g. instance segmentation. Specifically, QAM improves Mask R-CNN by  1 proposed QAM and QANet, our overall system wins 1st place in CVPR2019 COCO DensePose Challenge, and 1st place in Track 1 2 of CVPR2020 LIP Challenge. Code and models are available at https://github.com/soeaver/QANet.

READ FULL TEXT

page 1

page 3

page 4

page 8

research
03/01/2019

Mask Scoring R-CNN

Letting a deep network be aware of the quality of its own predictions is...
research
09/17/2018

Devil in the Details: Towards Accurate Single and Multiple Human Parsing

Human parsing has received considerable interest due to its wide applica...
research
10/22/2019

Self-Correction for Human Parsing

Labeling pixel-level masks for fine-grained semantic segmentation tasks,...
research
03/09/2015

Deep Human Parsing with Active Template Regression

In this work, the human parsing task, namely decomposing a human image i...
research
11/30/2018

Parsing R-CNN for Instance-Level Human Analysis

Instance-level human analysis is common in real-life scenarios and has m...
research
06/11/2021

Part-aware Panoptic Segmentation

In this work, we introduce the new scene understanding task of Part-awar...
research
06/14/2021

3rd Place Solution for Short-video Face Parsing Challenge

Short videos have many applications on fashion trends, hot spots, street...

Please sign up or login with your details

Forgot password? Click here to reset