Large Kernel Matters -- Improve Semantic Segmentation by Global Convolutional Network

Abstract

One of recent trends [30, 31, 14] in network architec- ture design is stacking small filters (e.g., 1x1 or 3x3) in the entire network because the stacked small filters is more ef- ficient than a large kernel, given the same computational complexity. However, in the field of semantic segmenta- tion, where we need to perform dense per-pixel prediction, we find that the large kernel (and effective receptive field) plays an important role when we have to perform the clas- sification and localization tasks simultaneously. Following our design principle, we propose a Global Convolutional Network to address both the classification and localization issues for the semantic segmentation. We also suggest a residual-based boundary refinement to further refine the ob- ject boundaries. Our approach achieves state-of-art perfor- mance on two public benchmarks and significantly outper- forms previous results, 82.2% (vs 80.2%) on PASCAL VOC 2012 dataset and 76.9% (vs 71.8%) on Cityscapes dataset.

References in corpus (1)

• Fast, Exact and Multi-Scale Inference for Semantic Image Segmentation with Deep Gaussian CRFs

Cited by in corpus (19)

• ShuffleNet V2: Practical Guidelines for Efficient CNN Architecture Design
• Pyramid Attention Network for Semantic Segmentation
• One-Shot Instance Segmentation
• Combining Background Subtraction Algorithms with Convolutional Neural Network
• S-Net: A Scalable Convolutional Neural Network for JPEG Compression Artifact Reduction
• SAN: Scale-Aware Network for Semantic Segmentation of High-Resolution Aerial Images
• 3D Global Convolutional Adversarial Network\\ for Prostate MR Volume Segmentation
• Fully Point-wise Convolutional Neural Network for Modeling Statistical Regularities in Natural Images
• Disentangled Image Matting
• Splenomegaly Segmentation using Global Convolutional Kernels and Conditional Generative Adversarial Networks
• IVUS-Net: An Intravascular Ultrasound Segmentation Network
• Pyramid Network with Online Hard Example Mining for Accurate Left Atrium Segmentation
• A Network Structure to Explicitly Reduce Confusion Errors in Semantic Segmentation
• Consensus Feature Network for Scene Parsing
• Rethinking Convolutional Semantic Segmentation Learning
• Dense Dilated Convolutions Merging Network for Semantic Mapping of Remote Sensing Images
• DFNet: Semantic Segmentation on Panoramic Images with Dynamic Loss Weights and Residual Fusion Block
• Splenomegaly Segmentation on Multi-modal MRI using Deep Convolutional Networks
• Semantic Human Matting