Back to Search Start Over

Learning to Predict Context-adaptive Convolution for Semantic Segmentation

Authors :
Liu, Jianbo
He, Junjun
Ren, Jimmy S.
Qiao, Yu
Li, Hongsheng
Publication Year :
2020

Abstract

Long-range contextual information is essential for achieving high-performance semantic segmentation. Previous feature re-weighting methods demonstrate that using global context for re-weighting feature channels can effectively improve the accuracy of semantic segmentation. However, the globally-sharing feature re-weighting vector might not be optimal for regions of different classes in the input image. In this paper, we propose a Context-adaptive Convolution Network (CaC-Net) to predict a spatially-varying feature weighting vector for each spatial location of the semantic feature maps. In CaC-Net, a set of context-adaptive convolution kernels are predicted from the global contextual information in a parameter-efficient manner. When used for convolution with the semantic feature maps, the predicted convolutional kernels can generate the spatially-varying feature weighting factors capturing both global and local contextual information. Comprehensive experimental results show that our CaC-Net achieves superior segmentation performance on three public datasets, PASCAL Context, PASCAL VOC 2012 and ADE20K.<br />Comment: Accepted in ECCV 2020

Details

Database :
arXiv
Publication Type :
Report
Accession number :
edsarx.2004.08222
Document Type :
Working Paper