Learning a Discriminative Feature Network for Semantic Segmentation

Abstract

Most existing methods of semantic segmentation still suffer from two aspects of challenges: intra-class inconsistency and inter-class indistinction. To tackle these two problems, we propose a Discriminative Feature Network (DFN), which contains two sub-networks: Smooth Network and Border Network. Specifically, to handle the intra-class inconsistency problem, we specially design a Smooth Network with Channel Attention Block and global average pooling to select the more discriminative features. Furthermore, we propose a Border Network to make the bilateral features of boundary distinguishable with deep semantic boundary supervision. Based on our proposed DFN, we achieve state-of-the-art performance 86.2% mean IOU on PASCAL VOC 2012 and 80.3% mean IOU on Cityscapes dataset.

arXiv commentAccepted to CVPR 2018. 10 pages, 9 figures

References in corpus (19)

• Rethinking Atrous Convolution for Semantic Image Segmentation
• Deep Residual Learning for Image Recognition
• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Multi-Scale Context Aggregation by Dilated Convolutions
• ParseNet: Looking Wider to See Better
• Recurrent Models of Visual Attention
• The Cityscapes Dataset for Semantic Urban Scene Understanding
• Fully Convolutional Networks for Semantic Segmentation
• Richer Convolutional Features for Edge Detection
• DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
• Pyramid Scene Parsing Network
• Semantic Image Segmentation via Deep Parsing Network
• Attention to Scale: Scale-aware Semantic Image Segmentation
• Understanding Convolution for Semantic Segmentation
• Object Contour Detection with a Fully Convolutional Encoder-Decoder Network
• Feedforward semantic segmentation with zoom-out features
• SCA-CNN: Spatial and Channel-wise Attention in Convolutional Networks for Image Captioning
• CASENet: Deep Category-Aware Semantic Edge Detection

Cited by in corpus (25)

• SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
• OCNet: Object Context Network for Scene Parsing
• Pyramid Attention Network for Semantic Segmentation
• Polarized Self-Attention: Towards High-quality Pixel-wise Regression
• BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
• CGNet: A Light-weight Context Guided Network for Semantic Segmentation
• DFANet: Deep Feature Aggregation for Real-Time Semantic Segmentation
• Global Aggregation then Local Distribution in Fully Convolutional Networks
• Strip Pooling: Rethinking Spatial Pooling for Scene Parsing
• CycleSegNet: Object Co-segmentation with Cycle Refinement and Region Correspondence
• Attention-guided Unified Network for Panoptic Segmentation
• Boundary-Aware Network for Fast and High-Accuracy Portrait Segmentation
• Joint Iris Segmentation and Localization Using Deep Multi-task Learning Framework
• Deep attention-based classification network for robust depth prediction
• RASNet: Segmentation for Tracking Surgical Instruments in Surgical Videos Using Refined Attention Segmentation Network
• Semantic Correlation Promoted Shape-Variant Context for Segmentation
• SegSort: Segmentation by Discriminative Sorting of Segments
• Tree-structured Kronecker Convolutional Network for Semantic Segmentation
• Triply Supervised Decoder Networks for Joint Detection and Segmentation
• OGNet: Salient Object Detection with Output-guided Attention Module
• Discriminative Online Learning for Fast Video Object Segmentation
• Multi Receptive Field Network for Semantic Segmentation
• SparseMask: Differentiable Connectivity Learning for Dense Image Prediction
• Hierarchical Pyramid Representations for Semantic Segmentation
• CNN-based Semantic Segmentation using Level Set Loss