Stacked Deconvolutional Network for Semantic Segmentation

Abstract

Recent progress in semantic segmentation has been driven by improving the spatial resolution under Fully Convolutional Networks (FCNs). To address this problem, we propose a Stacked Deconvolutional Network (SDN) for semantic segmentation. In SDN, multiple shallow deconvolutional networks, which are called as SDN units, are stacked one by one to integrate contextual information and guarantee the fine recovery of localization information. Meanwhile, inter-unit and intra-unit connections are designed to assist network training and enhance feature fusion since the connections improve the flow of information and gradient propagation throughout the network. Besides, hierarchical supervision is applied during the upsampling process of each SDN unit, which guarantees the discrimination of feature representations and benefits the network optimization. We carry out comprehensive experiments and achieve the new state-of-the-art results on three datasets, including PASCAL VOC 2012, CamVid, GATECH. In particular, our best model without CRF post-processing achieves an intersection-over-union score of 86.6% in the test set.

References in corpus (12)

• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• Rethinking Atrous Convolution for Semantic Image Segmentation
• Caffe: Convolutional Architecture for Fast Feature Embedding
• DeepPose: Human Pose Estimation via Deep Neural Networks
• ParseNet: Looking Wider to See Better
• Learning Deconvolution Network for Semantic Segmentation
• Stacked Hourglass Networks for Human Pose Estimation
• Deformable Convolutional Networks
• Wider or Deeper: Revisiting the ResNet Model for Visual Recognition
• Semantic Image Segmentation via Deep Parsing Network
• The One Hundred Layers Tiramisu: Fully Convolutional DenseNets for Semantic Segmentation
• Geometric Context from Videos

Cited by in corpus (25)

• Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation
• Deep High-Resolution Representation Learning for Visual Recognition
• A Survey on Deep Learning-based Architectures for Semantic Segmentation on 2D images
• FastFCN: Rethinking Dilated Convolution in the Backbone for Semantic Segmentation
• Dual Attention Network for Scene Segmentation
• DeeperLab: Single-Shot Image Parser
• Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
• Crowd Counting and Density Estimation by Trellis Encoder-Decoder Network
• Pyramid Stereo Matching Network
• ExFuse: Enhancing Feature Fusion for Semantic Segmentation
• A Survey on Deep Learning Methods for Robot Vision
• Deeply Self-Supervised Contour Embedded Neural Network Applied to Liver Segmentation
• Stacked U-Nets: A No-Frills Approach to Natural Image Segmentation
• Efficient Dense Modules of Asymmetric Convolution for Real-Time Semantic Segmentation
• Joint Semantic Segmentation and Boundary Detection using Iterative Pyramid Contexts
• SegSort: Segmentation by Discriminative Sorting of Segments
• Decoders Matter for Semantic Segmentation: Data-Dependent Decoding Enables Flexible Feature Aggregation
• DSNet: An Efficient CNN for Road Scene Segmentation
• TW-SMNet: Deep Multitask Learning of Tele-Wide Stereo Matching
• Multi-scale Attention U-Net (MsAUNet): A Modified U-Net Architecture for Scene Segmentation
• EfficientHRNet: Efficient Scaling for Lightweight High-Resolution Multi-Person Pose Estimation
• Learning Fully Dense Neural Networks for Image Semantic Segmentation
• Multi Receptive Field Network for Semantic Segmentation
• Adaptive Context Network for Scene Parsing
• Affinity Derivation and Graph Merge for Instance Segmentation