Guided Upsampling Network for Real-Time Semantic Segmentation

Abstract

Semantic segmentation architectures are mainly built upon an encoder-decoder structure. These models perform subsequent downsampling operations in the encoder. Since operations on high-resolution activation maps are computationally expensive, usually the decoder produces output segmentation maps by upsampling with parameters-free operators like bilinear or nearest-neighbor. We propose a Neural Network named Guided Upsampling Network which consists of a multiresolution architecture that jointly exploits high-resolution and large context information. Then we introduce a new module named Guided Upsampling Module (GUM) that enriches upsampling operators by introducing a learnable transformation for semantic maps. It can be plugged into any existing encoder-decoder architecture with little modifications and low additional computation cost. We show with quantitative and qualitative experiments how our network benefits from the use of GUM module. A comprehensive set of experiments on the publicly available Cityscapes dataset demonstrates that Guided Upsampling Network can efficiently process high-resolution images in real-time while attaining state-of-the art performances.

arXiv commentAccepted at BMVC 2018

References in corpus (8)

• Rethinking Atrous Convolution for Semantic Image Segmentation
• Semantic Image Segmentation with Deep Convolutional Nets and Fully Connected CRFs
• Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials
• ENet: A Deep Neural Network Architecture for Real-Time Semantic Segmentation
• Wider or Deeper: Revisiting the ResNet Model for Visual Recognition
• Simultaneous Detection and Segmentation
• Dilated Residual Networks
• ICNet for Real-Time Semantic Segmentation on High-Resolution Images

Cited by in corpus (10)

• Fast-SCNN: Fast Semantic Segmentation Network
• CARAFE: Content-Aware ReAssembly of FEatures
• SegBlocks: Block-Based Dynamic Resolution Networks for Real-Time Segmentation
• Real-time CNN-based Segmentation Architecture for Ball Detection in a Single View Setup
• SegFix: Model-Agnostic Boundary Refinement for Segmentation
• CARAFE++: Unified Content-Aware ReAssembly of FEatures
• Content-Aware Inter-Scale Cost Aggregation for Stereo Matching
• ShelfNet for Fast Semantic Segmentation
• FaPN: Feature-aligned Pyramid Network for Dense Image Prediction
• Collaborative Global-Local Networks for Memory-Efficient Segmentation of Ultra-High Resolution Images