Cited 0 time in webofscience Cited 0 time in scopus

Weakly Supervised Learning with Deep Convolutional Neural Networks for Semantic Segmentation Understanding semantic layout of images with minimum human supervision

Title
Weakly Supervised Learning with Deep Convolutional Neural Networks for Semantic Segmentation Understanding semantic layout of images with minimum human supervision
Authors
Hong, SeunghoonKwak, Su HaHan, Bohyung
DGIST Authors
Kwak, Su Ha
Issue Date
2017-11
Citation
IEEE Signal Processing Magazine, 34(6), 39-49
Type
Article
Article Type
Article
Keywords
Annotated Training DataClassification (of Information)ConvolutionConvolutional Neural NetworkDeep Neural NetworksImage SegmentationImprove PerformanceLarge Scale Data SetsNeural NetworksObject BoundariesObject DetectionSemantic SegmentationSemantic WebSemanticsVisual RecognitionWeakly Supervised Learning
ISSN
1053-5888
Abstract
Semantic segmentation is a popular visual recognition task whose goal is to estimate pixel-level object class labels in images. This problem has been recently handled by deep convolutional neural networks (DCNNs), and the state-of-theart techniques achieve impressive records on public benchmark data sets. However, learning DCNNs demand a large number of annotated training data while segmentation annotations in existing data sets are significantly limited in terms of both quantity and diversity due to the heavy annotation cost. Weakly supervised approaches tackle this issue by leveraging weak annotations such as image-level labels and bounding boxes, which are either readily available in existing large-scale data sets for image classification and object detection or easily obtained thanks to their low annotation costs. The main challenge in weakly supervised semantic segmentation then is the incomplete annotations that miss accurate object boundary information required to learn segmentation. This article provides a comprehensive overview of weakly supervised approaches for semantic segmentation. Specifically, we describe how the approaches overcome the limitations and discuss research directions worthy of investigation to improve performance. © 1991-2012 IEEE.
URI
http://hdl.handle.net/20.500.11750/4841
DOI
10.1109/MSP.2017.2742558
Publisher
Institute of Electrical and Electronics Engineers Inc.
Files:
There are no files associated with this item.
Collection:
Department of Information and Communication EngineeringVision and Learning Lab(VL Lab)1. Journal Articles


qrcode mendeley

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.

BROWSE