Dilated Residual Networks

Abstract

Convolutional networks for image classification progressively reduce resolution until the image is represented by tiny feature maps in which the spatial structure of the scene is no longer discernible. Such loss of spatial acuity can limit image classification accuracy and complicate the transfer of the model to downstream applications that require detailed scene understanding. These problems can be alleviated by dilation, which increases the resolution of output feature maps without reducing the receptive field of individual neurons. We show that dilated residual networks (DRNs) outperform their non-dilated counterparts in image classification without increasing the model's depth or complexity. We then study gridding artifacts introduced by dilation, develop an approach to removing these artifacts (`degridding'), and show that this further increases the performance of DRNs. In addition, we show that the accuracy advantage of DRNs is further magnified in downstream applications such as object localization and semantic segmentation.

Published at the Conference on Computer Vision and Pattern Recognition (CVPR 2017)

References in corpus (2)

• Going Deeper with Convolutions
• Learning Deconvolution Network for Semantic Segmentation

Cited by in corpus (12)

• High-Resolution Representations for Labeling Pixels and Regions
• CyCADA: Cycle-Consistent Adversarial Domain Adaptation
• Image to Image Translation for Domain Adaptation
• Significance-aware Information Bottleneck for Domain Adaptive Semantic Segmentation
• Attention-based Context Aggregation Network for Monocular Depth Estimation
• RegNet: Learning the Optimization of Direct Image-to-Image Pose Registration
• Accurate Automatic Segmentation of Amygdala Subnuclei and Modeling of Uncertainty via Bayesian Fully Convolutional Neural Network
• Context-Constrained Accurate Contour Extraction for Occlusion Edge Detection
• Multichannel Semantic Segmentation with Unsupervised Domain Adaptation
• Appearance-Based Gaze Estimation Using Dilated-Convolutions
• Residual Pyramid FCN for Robust Follicle Segmentation
• Holistic Decomposition Convolution for Effective Semantic Segmentation of 3D MR Images