Decoupled Deep Neural Network for Semi-supervised Semantic Segmentation

Abstract

We propose a novel deep neural network architecture for semi-supervised semantic segmentation using heterogeneous annotations. Contrary to existing approaches posing semantic segmentation as a single task of region-based classification, our algorithm decouples classification and segmentation, and learns a separate network for each task. In this architecture, labels associated with an image are identified by classification network, and binary segmentation is subsequently performed for each identified label in segmentation network. The decoupled architecture enables us to learn classification and segmentation networks separately based on the training data with image-level and pixel-wise class labels, respectively. It facilitates to reduce search space for segmentation effectively by exploiting class-specific activation maps obtained from bridging layers. Our algorithm shows outstanding performance compared to other semi-supervised approaches even with much less training images with strong annotations in PASCAL VOC dataset.

Added a link to the project page for more comprehensive illustration of results

References in corpus (2)

• Conditional Random Fields as Recurrent Neural Networks
• Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation

Cited by in corpus (25)

• Low-Dose CT with a Residual Encoder-Decoder Convolutional Neural Network (RED-CNN)
• Fully Convolutional Networks for Semantic Segmentation
• STFCN: Spatio-Temporal FCN for Semantic Video Segmentation
• Unsupervised Domain Adaptation in Semantic Segmentation: a Review
• S-Net: A Scalable Convolutional Neural Network for JPEG Compression Artifact Reduction
• Budget-aware Semi-Supervised Semantic and Instance Segmentation
• Evaluating Weakly Supervised Object Localization Methods Right
• From handcrafted to deep local features
• Certifiably Robust Interpretation in Deep Learning
• Deeply Supervised Active Learning for Finger Bones Segmentation
• Universal Semi-Supervised Semantic Segmentation
• Naive-Student: Leveraging Semi-Supervised Learning in Video Sequences for Urban Scene Segmentation
• Unsupervised Medical Image Segmentation with Adversarial Networks: From Edge Diagrams to Segmentation Maps
• Explainable Object-induced Action Decision for Autonomous Vehicles
• Are We Hungry for 3D LiDAR Data for Semantic Segmentation? A Survey and Experimental Study
• The iMaterialist Fashion Attribute Dataset
• TriangleNet: Edge Prior Augmented Network for Semantic Segmentation through Cross-Task Consistency
• Multiview Cross-supervision for Semantic Segmentation
• TricycleGAN: Unsupervised Image Synthesis and Segmentation Based on Shape Priors
• Construct Deep Neural Networks Based on Direct Sampling Methods for Solving Electrical Impedance Tomography
• Automatic Labeled LiDAR Data Generation based on Precise Human Model
• Attentive Deep Regression Networks for Real-Time Visual Face Tracking in Video Surveillance
• Hierarchical Deep Network with Uncertainty-aware Semi-supervised Learning for Vessel Segmentation
• DASNet: Reducing Pixel-level Annotations for Instance and Semantic Segmentation
• Iterative Self-Learning: Semi-Supervised Improvement to Dataset Volumes and Model Accuracy