Fully Convolutional Multi-Class Multiple Instance Learning

Abstract

Multiple instance learning (MIL) can reduce the need for costly annotation in tasks such as semantic segmentation by weakening the required degree of supervision. We propose a novel MIL formulation of multi-class semantic segmentation learning by a fully convolutional network. In this setting, we seek to learn a semantic segmentation model from just weak image-level labels. The model is trained end-to-end to jointly optimize the representation while disambiguating the pixel-image label assignment. Fully convolutional training accepts inputs of any size, does not need object proposal pre-processing, and offers a pixelwise loss map for selecting latent instances. Our multi-class MIL loss exploits the further supervision given by images with multiple labels. We evaluate this approach through preliminary experiments on the PASCAL VOC segmentation challenge.

in ICLR 2015

References in corpus (4)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• Fully Convolutional Networks for Semantic Segmentation
• A Tractable Inference Algorithm for Diagnosing Multiple Diseases
• Detector Discovery in the Wild: Joint Multiple Instance and Representation Learning

Cited by in corpus (70)

• FCNs in the Wild: Pixel-level Adversarial and Constraint-based Adaptation
• STC: A Simple to Complex Framework for Weakly-supervised Semantic Segmentation
• Classifying and Segmenting Microscopy Images Using Convolutional Multiple Instance Learning
• Adversarial Learning for Semi-Supervised Semantic Segmentation
• Mixed supervision for surface-defect detection: from weakly to fully supervised learning
• Weakly- and Semi-Supervised Learning of a DCNN for Semantic Image Segmentation
• Constrained-CNN losses for weakly supervised segmentation
• Monte-Carlo Sampling applied to Multiple Instance Learning for Histological Image Classification
• Can Image-Level Labels Replace Pixel-Level Labels for Image Parsing
• U-Net: Convolutional Networks for Biomedical Image Segmentation
• A Comprehensive Analysis of Weakly-Supervised Semantic Segmentation in Different Image Domains
• Weakly Supervised Medical Diagnosis and Localization from Multiple Resolutions
• Webly Supervised Learning of Convolutional Networks
• Fully Convolutional Adaptation Networks for Semantic Segmentation
• Gated CRF Loss for Weakly Supervised Semantic Image Segmentation
• Semi and Weakly Supervised Semantic Segmentation Using Generative Adversarial Network
• Prior-aware Neural Network for Partially-Supervised Multi-Organ Segmentation
• What's the Point: Semantic Segmentation with Point Supervision
• Object Region Mining with Adversarial Erasing: A Simple Classification to Semantic Segmentation Approach
• Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi- Supervised Semantic Segmentation
• Incorporating Network Built-in Priors in Weakly-supervised Semantic Segmentation
• Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Features
• Patch-based Convolutional Neural Network for Whole Slide Tissue Image Classification
• Simple Does It: Weakly Supervised Instance and Semantic Segmentation
• Exploiting saliency for object segmentation from image level labels
• Weakly Supervised Semantic Segmentation using Web-Crawled Videos
• Weakly Supervised Semantic Point Cloud Segmentation:Towards 10X Fewer Labels
• Built-in Foreground/Background Prior for Weakly-Supervised Semantic Segmentation
• Unsupervised Domain Adaptation in Semantic Segmentation: a Review
• Applying Topological Persistence in Convolutional Neural Network for Music Audio Signals
• Weakly-supervised detection of AMD-related lesions in color fundus images using explainable deep learning
• Budget-aware Semi-Supervised Semantic and Instance Segmentation
• No More Discrimination: Cross City Adaptation of Road Scene Segmenters
• Effective Use of Synthetic Data for Urban Scene Semantic Segmentation
• Deep Extreme Cut: From Extreme Points to Object Segmentation
• Every Annotation Counts: Multi-label Deep Supervision for Medical Image Segmentation
• Deconvolutional Feature Stacking for Weakly-Supervised Semantic Segmentation
• Recent Advances in the Applications of Convolutional Neural Networks to Medical Image Contour Detection
• PCAMs: Weakly Supervised Semantic Segmentation Using Point Supervision
• Learning Transferrable Knowledge for Semantic Segmentation with Deep Convolutional Neural Network
• Decoupled Spatial Neural Attention for Weakly Supervised Semantic Segmentation
• Convolutional Simplex Projection Network (CSPN) for Weakly Supervised Semantic Segmentation
• NuClick: A Deep Learning Framework for Interactive Segmentation of Microscopy Images
• DeepCut: Object Segmentation from Bounding Box Annotations using Convolutional Neural Networks
• ProNet: Learning to Propose Object-specific Boxes for Cascaded Neural Networks
• Weakly and Semi Supervised Human Body Part Parsing via Pose-Guided Knowledge Transfer
• Deep Structured Scene Parsing by Learning with Image Descriptions
• Large-Scale 3D Scene Classification With Multi-View Volumetric CNN
• Learning Semantic Segmentation with Diverse Supervision
• ScribbleSup: Scribble-Supervised Convolutional Networks for Semantic Segmentation
• Block Annotation: Better Image Annotation for Semantic Segmentation with Sub-Image Decomposition
• Are We Hungry for 3D LiDAR Data for Semantic Segmentation? A Survey and Experimental Study
• Scoring Root Necrosis in Cassava Using Semantic Segmentation
• Specialized Decision Surface and Disentangled Feature for Weakly-Supervised Polyphonic Sound Event Detection
• Keypoint Based Weakly Supervised Human Parsing
• Diverse Sampling for Self-Supervised Learning of Semantic Segmentation
• Unsupervised Learning of Important Objects from First-Person Videos
• Constrained Convolutional Neural Networks for Weakly Supervised Segmentation
• Weakly Supervised Image Annotation and Segmentation with Objects and Attributes
• Generating Self-Guided Dense Annotations for Weakly Supervised Semantic Segmentation
• Learning Human Poses from Actions
• Automatic Image Labelling at Pixel Level
• Weakly supervised training of pixel resolution segmentation models on whole slide images
• Closed-Loop Adaptation for Weakly-Supervised Semantic Segmentation
• Learning Pixel-wise Labeling from the Internet without Human Interaction
• Hierarchical Scene Parsing by Weakly Supervised Learning with Image Descriptions
• VALUE: Large Scale Voting-based Automatic Labelling for Urban Environments
• Seed, Expand and Constrain: Three Principles for Weakly-Supervised Image Segmentation
• Differentiable Projection for Constrained Deep Learning
• Learning to segment with image-level supervision