Semantic Segmentation using Adversarial Networks

Abstract

Adversarial training has been shown to produce state of the art results for generative image modeling. In this paper we propose an adversarial training approach to train semantic segmentation models. We train a convolutional semantic segmentation network along with an adversarial network that discriminates segmentation maps coming either from the ground truth or from the segmentation network. The motivation for our approach is that it can detect and correct higher-order inconsistencies between ground truth segmentation maps and the ones produced by the segmentation net. Our experiments show that our adversarial training approach leads to improved accuracy on the Stanford Background and PASCAL VOC 2012 datasets.

Cited by in corpus (95)

• Medical Image Segmentation Using Deep Learning: A Survey
• Anatomically Constrained Neural Networks (ACNN): Application to Cardiac Image Enhancement and Segmentation
• Evolution of Image Segmentation using Deep Convolutional Neural Network: A Survey
• Generative Adversarial Networks in Computer Vision: A Survey and Taxonomy
• Learning Adversarially Fair and Transferable Representations
• Monocular Semantic Occupancy Grid Mapping with Convolutional Variational Encoder-Decoder Networks
• DE-GAN: A Conditional Generative Adversarial Network for Document Enhancement
• An Introduction to Image Synthesis with Generative Adversarial Nets
• Attention U-Net Based Adversarial Architectures for Chest X-ray Lung Segmentation
• Machine learning based hyperspectral image analysis: A survey
• Deep and Statistical Learning in Biomedical Imaging: State of the Art in 3D MRI Brain Tumor Segmentation
• Data synthesis and adversarial networks: A review and meta-analysis in cancer imaging
• Towards Goal-Oriented Semantic Signal Processing: Applications and Future Challenges
• Self-paced and self-consistent co-training for semi-supervised image segmentation
• Multiple-Human Parsing in the Wild
• Leveraging Uncertainty Estimates for Predicting Segmentation Quality
• Lightweight Modules for Efficient Deep Learning based Image Restoration
• Translating and Segmenting Multimodal Medical Volumes with Cycle- and Shape-Consistency Generative Adversarial Network
• A Comprehensive Review of Modern Object Segmentation Approaches
• Rethinking Monocular Depth Estimation with Adversarial Training
• Robust Iris Segmentation Based on Fully Convolutional Networks and Generative Adversarial Networks
• Predicting Deeper into the Future of Semantic Segmentation
• Conditional Generation of Medical Images via Disentangled Adversarial Inference
• AGRNet: Adaptive Graph Representation Learning and Reasoning for Face Parsing
• Speaking the Same Language: Matching Machine to Human Captions by Adversarial Training
• Improving Lesion Segmentation for Diabetic Retinopathy using Adversarial Learning
• SCAN: Structure Correcting Adversarial Network for Organ Segmentation in Chest X-rays
• Brain MRI Tumor Segmentation with Adversarial Networks
• Generative adversarial networks and adversarial methods in biomedical image analysis
• Deep Learning in Medical Ultrasound Image Segmentation: a Review
• Toward Simulating Environments in Reinforcement Learning Based Recommendations
• DatasetGAN: Efficient Labeled Data Factory with Minimal Human Effort
• Structured Knowledge Distillation for Dense Prediction
• Correlation Maximized Structural Similarity Loss for Semantic Segmentation
• Deep Adversarial Attention Alignment for Unsupervised Domain Adaptation: the Benefit of Target Expectation Maximization
• Adaptive Affinity Fields for Semantic Segmentation
• LiDARNet: A Boundary-Aware Domain Adaptation Model for Point Cloud Semantic Segmentation
• SegDiff: Image Segmentation with Diffusion Probabilistic Models
• Deep Co-Training for Semi-Supervised Image Segmentation
• Pulsar Candidate Identification with Artificial Intelligence Techniques
• Improving Neural Machine Translation with Conditional Sequence Generative Adversarial Nets
• A Survey on Deep Learning Methods for Semantic Image Segmentation in Real-Time
• 3D Global Convolutional Adversarial Network\\ for Prostate MR Volume Segmentation
• Hierarchical Human Parsing with Typed Part-Relation Reasoning
• Semantic Segmentation for Partially Occluded Apple Trees Based on Deep Learning
• Adversarial Loss for Semantic Segmentation of Aerial Imagery
• Advances in Medical Image Segmentation: A Comprehensive Survey with a Focus on Lumbar Spine Applications
• Face Completion with Semantic Knowledge and Collaborative Adversarial Learning
• Fast video object segmentation with Spatio-Temporal GANs
• GestureGAN for Hand Gesture-to-Gesture Translation in the Wild
• Deep Cross-Modal Audio-Visual Generation
• Recent Advances in Fibrosis and Scar Segmentation from Cardiac MRI: A State-of-the-Art Review and Future Perspectives
• Secure Domain Adaptation with Multiple Sources
• AMIL: Adversarial Multi Instance Learning for Human Pose Estimation
• Deep Multi-Facial Patches Aggregation Network For Facial Expression Recognition
• Uncertainty-Driven Semantic Segmentation through Human-Machine Collaborative Learning
• Judge the Judges: A Large-Scale Evaluation Study of Neural Language Models for Online Review Generation
• Improving Consistency and Correctness of Sequence Inpainting using Semantically Guided Generative Adversarial Network
• Supervised Adversarial Networks for Image Saliency Detection
• Semi-Supervised Segmentation of Concrete Aggregate Using Consensus Regularisation and Prior Guidance
• Exploration of Convolutional Neural Network Architectures for Large Region Map Automation
• Adversarial Convolutional Networks with Weak Domain-Transfer for Multi-Sequence Cardiac MR Images Segmentation
• SDE approximations of GANs training and its long-run behavior
• 3D Semantic Scene Completion from a Single Depth Image using Adversarial Training
• Semi-supervised Semantic Segmentation of Prostate and Organs-at-Risk on 3D Pelvic CT Images
• Generative Smoke Removal
• Comparing Adversarial and Supervised Learning for Organs at Risk Segmentation in CT images
• Learning Discriminators as Energy Networks in Adversarial Learning
• Medical Image Segmentation with Limited Supervision: A Review of Deep Network Models
• Adversarial Learning for Image Forensics Deep Matching with Atrous Convolution
• TLGAN: document Text Localization using Generative Adversarial Nets
• Points2Polygons: Context-Based Segmentation from Weak Labels Using Adversarial Networks
• EMIXER: End-to-end Multimodal X-ray Generation via Self-supervision
• Weakly-supervised Generative Adversarial Networks for medical image classification
• Phase Collaborative Network for Two-Phase Medical Image Segmentation
• Generator evaluator-selector net for panoptic image segmentation and splitting unfamiliar objects into parts
• Edge-competing Pathological Liver Vessel Segmentation with Limited Labels
• High Diversity Attribute Guided Face Generation with GANs
• Visual Boundary Knowledge Translation for Foreground Segmentation
• Image Embedded Segmentation: Uniting Supervised and Unsupervised Objectives for Segmenting Histopathological Images
• SUSAN: Segment Unannotated image Structure using Adversarial Network
• When Autonomous Systems Meet Accuracy and Transferability through AI: A Survey
• Automatic Liver Segmentation with Adversarial Loss and Convolutional Neural Network
• Splenomegaly Segmentation on Multi-modal MRI using Deep Convolutional Networks
• Convolutional generative adversarial imputation networks for spatio-temporal missing data in storm surge simulations
• Coarse-to-Fine Gaze Redirection with Numerical and Pictorial Guidance
• DETCID: Detection of Elongated Touching Cells with Inhomogeneous Illumination using a Deep Adversarial Network
• \emph{cm}SalGAN: RGB-D Salient Object Detection with Cross-View Generative Adversarial Networks
• Adversarial Structure Matching for Structured Prediction Tasks
• Synthetic Convolutional Features for Improved Semantic Segmentation
• Minimizing Labeling Effort for Tree Skeleton Segmentation using an Automated Iterative Training Methodology
• EnMcGAN: Adversarial Ensemble Learning for 3D Complete Renal Structures Segmentation
• Robustness-aware 2-bit quantization with real-time performance for neural network
• Towards Robust Partially Supervised Multi-Structure Medical Image Segmentation on Small-Scale Data
• Generative Adversarial Network: Some Analytical Perspectives