Weakly-and Semi-Supervised Learning of a Deep Convolutional Network for Semantic Image Segmentation

Abstract

Deep convolutional neural networks (DCNNs) trained on a large number of images with strong pixel-level annotations have recently significantly pushed the state-of-art in semantic image segmentation. We study the more challenging problem of learning DCNNs for semantic image segmentation from either (1) weakly annotated training data such as bounding boxes or image-level labels or (2) a combination of few strongly labeled and many weakly labeled images, sourced from one or multiple datasets. We develop Expectation-Maximization (EM) methods for semantic image segmentation model training under these weakly supervised and semi-supervised settings. Extensive experimental evaluation shows that the proposed techniques can learn models delivering competitive results on the challenging PASCAL VOC 2012 image segmentation benchmark, while requiring significantly less annotation effort. We share source code implementing the proposed system at https: //bitbucket.org/deeplab/deeplab-public.

WILDCAT: Weakly Supervised Learning of Deep ConvNets for Image Classification, Pointwise Localization and Segmentation

“…Another important WSL application is segmentation. Many methods are based on MIL framework: MIL-FCN [49] extends MIL to multi-class segmentation, MIL-Base [50] introduces a soft extension of MIL, EM-Adapt [45] includes an adaptive bias into the MIL framework, and Constrained CNN (CCNN) [48] uses a loss function optimized for any set of linear constraints on the output space of a CNN.…”

High-Resolution Remote Sensing Data Classification over Urban Areas Using Random Forest Ensemble and Fully Connected Conditional Random Field

“…Specifically, the validation set consists of 3 tiles (areas: 26,28,30) and the testing set consists of 3 tiles (areas: 32, 34, 37). The remaining 10 tiles represent the training set, and each tile is used for training an individual RF classifier.…”

“…At the same time, progress in graphic processing unit (GPU) and parallel computing technology has significantly increased the computing capability, such that learning a more complicated classifier with a larger amount of training data has becomes accessible to more researchers. Specifically, one of the most successful practices in this direction was the launch of deep CNN (convolutional neural networks) [19][20][21][22] in the computer vision community, which has become the dominant method for visual recognition and semantic classification [13,[23][24][25][26]. Furthermore, one of the most distinct characteristics of CNN is its ability to automatically learn the most suitable features, which has made the manual feature extraction process that is used in the traditional supervised-based classification methods unnecessary.…”

A Survey on Machine Learning‐Based Mobile Big Data Analysis: Challenges and Applications

“…Subsampling layers, which are also called pooling layers, adjust outputs from convolutional layer to get translation invariance. CNN is mainly applied in computer vision field for big data, for example, image classification [124,125] and image segmentation [126]. Document (or textual) representation, also part of NLP, is the basic method for information retrieval and important to understand natural language.…”