Pyramid Scene Parsing Network

TL;DR

The paper introduces PSPNet, a novel neural network architecture utilizing pyramid pooling for scene parsing, achieving state-of-the-art accuracy on multiple datasets and winning several benchmarks.

Contribution

It proposes a pyramid pooling module within PSPNet that effectively captures global context for improved pixel-level scene parsing.

Findings

Achieved 85.4% mIoU on PASCAL VOC 2012

Achieved 80.2% accuracy on Cityscapes

Won ImageNet scene parsing challenge 2016

Abstract

Scene parsing is challenging for unrestricted open vocabulary and diverse scenes. In this paper, we exploit the capability of global context information by different-region-based context aggregation through our pyramid pooling module together with the proposed pyramid scene parsing network (PSPNet). Our global prior representation is effective to produce good quality results on the scene parsing task, while PSPNet provides a superior framework for pixel-level prediction tasks. The proposed approach achieves state-of-the-art performance on various datasets. It came first in ImageNet scene parsing challenge 2016, PASCAL VOC 2012 benchmark and Cityscapes benchmark. A single PSPNet yields new record of mIoU accuracy 85.4% on PASCAL VOC 2012 and accuracy 80.2% on Cityscapes.