Pyramid Scene Parsing Network

Abstract

Scene parsing is challenging for unrestricted open vocabulary and diverse scenes. In this paper, we exploit the capability of global context information by different-region-based context aggregation through our pyramid pooling module together with the proposed pyramid scene parsing network (PSPNet). Our global prior representation is effective to produce good quality results on the scene parsing task, while PSPNet provides a superior framework for pixel-level prediction tasks. The proposed approach achieves state-of-the-art performance on various datasets. It came first in ImageNet scene parsing challenge 2016, PASCAL VOC 2012 benchmark and Cityscapes benchmark. A single PSPNet yields new record of mIoU accuracy 85.4% on PASCAL VOC 2012 and accuracy 80.2% on Cityscapes.

arXiv commentCVPR 2017

References in corpus (21)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Deep Residual Learning for Image Recognition
• Caffe: Convolutional Architecture for Fast Feature Embedding
• Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition
• Multi-Scale Context Aggregation by Dilated Convolutions
• Going Deeper with Convolutions
• ParseNet: Looking Wider to See Better
• The Cityscapes Dataset for Semantic Urban Scene Understanding
• Fully Convolutional Networks for Semantic Segmentation
• Object Detectors Emerge in Deep Scene CNNs
• DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs
• Learning Deconvolution Network for Semantic Segmentation
• Towards Good Practices for Very Deep Two-Stream ConvNets
• Scalable, High-Quality Object Detection
• Semantic Understanding of Scenes through the ADE20K Dataset
• Semantic Image Segmentation via Deep Parsing Network
• Bridging Category-level and Instance-level Semantic Image Segmentation
• BoxSup: Exploiting Bounding Boxes to Supervise Convolutional Networks for Semantic Segmentation
• Attention to Scale: Scale-aware Semantic Image Segmentation
• Feedforward semantic segmentation with zoom-out features
• Relay Backpropagation for Effective Learning of Deep Convolutional Neural Networks

Cited by in corpus (120)

• Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics
• Fast-SCNN: Fast Semantic Segmentation Network
• Evolution of Image Segmentation using Deep Convolutional Neural Network: A Survey
• Semantic Labeling in Very High Resolution Images via a Self-Cascaded Convolutional Neural Network
• Direction-aware Spatial Context Features for Shadow Detection
• Deep Learning Enables Automatic Detection and Segmentation of Brain Metastases on Multi-Sequence MRI
• Bag of Tricks for Image Classification with Convolutional Neural Networks
• Context Encoding for Semantic Segmentation
• BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation
• 2018 Robotic Scene Segmentation Challenge
• Person Transfer GAN to Bridge Domain Gap for Person Re-Identification
• Automated Evaluation of Semantic Segmentation Robustness for Autonomous Driving
• Learning a Discriminative Feature Network for Semantic Segmentation
• Densely Connected Pyramid Dehazing Network
• Pyramid Stereo Matching Network
• PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing
• Improving Fast Segmentation With Teacher-student Learning
• ExFuse: Enhancing Feature Fusion for Semantic Segmentation
• Driving Scene Perception Network: Real-time Joint Detection, Depth Estimation and Semantic Segmentation
• Real-time Semantic Image Segmentation via Spatial Sparsity
• FISHING Net: Future Inference of Semantic Heatmaps In Grids
• Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network
• A Survey on Deep Learning Methods for Robot Vision
• CANet: Class-Agnostic Segmentation Networks with Iterative Refinement and Attentive Few-Shot Learning
• Fully Convolutional Networks for Chip-wise Defect Detection Employing Photoluminescence Images
• Relation-Aware Global Attention for Person Re-identification
• Deep Watershed Transform for Instance Segmentation
• Deep Layer Aggregation
• Sharing Residual Units Through Collective Tensor Factorization in Deep Neural Networks
• Autonomous Driving in Reality with Reinforcement Learning and Image Translation
• Instance-level Human Parsing via Part Grouping Network
• Improving Multi-Modal Learning with Uni-Modal Teachers
• Attention-based Context Aggregation Network for Monocular Depth Estimation
• In Defense of Classical Image Processing: Fast Depth Completion on the CPU
• Adapting Semantic Segmentation Models for Changes in Illumination and Camera Perspective
• Natural Language Guided Visual Relationship Detection
• DCAN: Dual Channel-wise Alignment Networks for Unsupervised Scene Adaptation
• LEGO: Learning Edge with Geometry all at Once by Watching Videos
• Knowledge Adaptation for Efficient Semantic Segmentation
• PaveSAM Segment Anything for Pavement Distress
• Agile Amulet: Real-Time Salient Object Detection with Contextual Attention
• All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation
• Close Yet Distinctive Domain Adaptation
• Effective Use of Dilated Convolutions for Segmenting Small Object Instances in Remote Sensing Imagery
• A Graph Theoretic Framework of Recomputation Algorithms for Memory-Efficient Backpropagation
• Graph-Based Global Reasoning Networks
• Scene Parsing with Global Context Embedding
• PlaneNet: Piece-wise Planar Reconstruction from a Single RGB Image
• Adaptive Semantic Segmentation with a Strategic Curriculum of Proxy Labels
• Deep Extreme Cut: From Extreme Points to Object Segmentation
• ACE: Adapting to Changing Environments for Semantic Segmentation
• V-NAS: Neural Architecture Search for Volumetric Medical Image Segmentation
• Learning Deep Representations for Semantic Image Parsing: a Comprehensive Overview
• Label-Driven Reconstruction for Domain Adaptation in Semantic Segmentation
• Distilling Knowledge via Knowledge Review
• Deep Online Video Stabilization
• Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation
• Points2Pix: 3D Point-Cloud to Image Translation using conditional Generative Adversarial Networks
• Semantic Segmentation of Human Thigh Quadriceps Muscle in Magnetic Resonance Images
• Adaptive Weighting Multi-Field-of-View CNN for Semantic Segmentation in Pathology
• Can we cover navigational perception needs of the visually impaired by panoptic segmentation?
• Do Normalization Layers in a Deep ConvNet Really Need to Be Distinct?
• Error Correction for Dense Semantic Image Labeling
• BRULÈ: Barycenter-Regularized Unsupervised Landmark Extraction
• ArbiText: Arbitrary-Oriented Text Detection in Unconstrained Scene
• SegStereo: Exploiting Semantic Information for Disparity Estimation
• DeLS-3D: Deep Localization and Segmentation with a 3D Semantic Map
• BusyHands: A Hand-Tool Interaction Database for Assembly Tasks Semantic Segmentation
• Value of Temporal Dynamics Information in Driving Scene Segmentation
• Sketch2code: Generating a website from a paper mockup
• Beyond Forward Shortcuts: Fully Convolutional Master-Slave Networks (MSNets) with Backward Skip Connections for Semantic Segmentation
• Distantly Supervised Road Segmentation
• CRAFT: Complementary Recommendations Using Adversarial Feature Transformer
• Video Object Segmentation with Joint Re-identification and Attention-Aware Mask Propagation
• Boundary-sensitive Network for Portrait Segmentation
• EfficientHRNet: Efficient Scaling for Lightweight High-Resolution Multi-Person Pose Estimation
• Multi-domain semantic segmentation with pyramidal fusion
• Learning with Hierarchical Complement Objective
• End-to-end Interpretable Neural Motion Planner
• Super-BPD: Super Boundary-to-Pixel Direction for Fast Image Segmentation
• Identifying Most Walkable Direction for Navigation in an Outdoor Environment
• Neuron-level Selective Context Aggregation for Scene Segmentation
• Dual Encoder Fusion U-Net (DEFU-Net) for Cross-manufacturer Chest X-ray Segmentation
• Bipartite Conditional Random Fields for Panoptic Segmentation
• Not Using the Car to See the Sidewalk: Quantifying and Controlling the Effects of Context in Classification and Segmentation
• Low-Latency Video Semantic Segmentation
• SurfConv: Bridging 3D and 2D Convolution for RGBD Images
• Rethinking Convolutional Semantic Segmentation Learning
• AttendSeg: A Tiny Attention Condenser Neural Network for Semantic Segmentation on the Edge
• A deep learning based multiscale approach to segment cancer area in liver whole slide image
• Automatic Real-time Background Cut for Portrait Videos
• Deep Consensus Learning
• YUVMultiNet: Real-time YUV multi-task CNN for autonomous driving
• A Survey of FPGA-Based Robotic Computing
• Feature-Fused Context-Encoding Network for Neuroanatomy Segmentation
• A Survey on Sensor Technologies for Unmanned Ground Vehicles
• Dynamic Video Segmentation Network
• Using UNet and PSPNet to explore the reusability principle of CNN parameters
• BABO: Background Activation Black-Out for Efficient Object Detection
• Scene Parsing via Dense Recurrent Neural Networks with Attentional Selection
• Overcoming Missing and Incomplete Modalities with Generative Adversarial Networks for Building Footprint Segmentation
• Exploring the parameter reusability of CNN
• Semantic Segmentation via Highly Fused Convolutional Network with Multiple Soft Cost Functions
• Dense Dilated Convolutions Merging Network for Semantic Mapping of Remote Sensing Images
• ESFNet: Efficient Network for Building Extraction from High-Resolution Aerial Images
• Split-Merge Pooling
• Using Cross-Model EgoSupervision to Learn Cooperative Basketball Intention
• Quality-Aware Network for Face Parsing
• Rethinking Fully Convolutional Networks for the Analysis of Photoluminescence Wafer Images
• A Study on Trees's Knots Prediction from their Bark Outer-Shape
• Is Each Layer Non-trivial in CNN?
• Per-Pixel Feedback for improving Semantic Segmentation
• Target-Tailored Source-Transformation for Scene Graph Generation
• A Survey On 3D Inner Structure Prediction from its Outer Shape
• DV3+HED+: A DCNNs-based Framework to Monitor Temporary Works and ESAs in Railway Construction Project Using VHR Satellite Images
• Multilayer Dense Connections for Hierarchical Concept Classification
• Learning Dilation Factors for Semantic Segmentation of Street Scenes
• Computationally efficient cardiac views projection using 3D Convolutional Neural Networks
• Exploring and Exploiting Diversity for Image Segmentation
• PSGAN++: Robust Detail-Preserving Makeup Transfer and Removal