Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided
碩士 === 國立中央大學 === 資訊工程學系 === 105 === In recent years, as a branch of machine learning, deep learning play an important role in Artificial Intelligence, which Convolutional Neural Network (CNN) has a breakthrough Performance in the image classification when comparing with traditional classification m...
Main Authors: | , |
---|---|
Other Authors: | |
Format: | Others |
Language: | zh-TW |
Published: |
2017
|
Online Access: | http://ndltd.ncl.edu.tw/handle/4vcn3f |
id |
ndltd-TW-105NCU05392140 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-TW-105NCU053921402019-05-16T00:08:09Z http://ndltd.ncl.edu.tw/handle/4vcn3f Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided 基於物件遮罩與邊界引導多尺度遞迴卷積神經 網路之語意分割 Kuan-Chung Wang 王冠中 碩士 國立中央大學 資訊工程學系 105 In recent years, as a branch of machine learning, deep learning play an important role in Artificial Intelligence, which Convolutional Neural Network (CNN) has a breakthrough Performance in the image classification when comparing with traditional classification methods. The emergence of the full Convolutional Network (FCN)[10] also makes the study of image semantic segmentation flourish. In contrast to past work clustering according to the image texture and color, FCN joined the training of semantic information to improve the accuracy of semantic segmentation. Our paper combines the advantages of two networks, an object boundary based approach to strengthen the integrity of edge and the object itself, and the other is responsible for the prediction of image semantic segmentation, proposed an end-to-end training network architecture. In this paper, proposed architecture improves the DT EdgeNet (Domain Transform with EdgeNet)[11]. Here, we combined the OBG-FCN [12] mask network and replaced the [11] edge network. The used mask network can predict background, object, and object edge reference diagrams. In addition, our architecture uses multi-scale ResNet-101 as the base network and introduces multi-scale Atrous Convolution to architecture training to preserve the dimensions of the feature map, which increases the receptive and further to enhance the accuracy of semantic segmentation. In the experiments, we got the high performance of recognition on the VOC2012 test set. In addition, we combined extraction of object bounding box generated by Faster RCNN and result of proposed semantic segmentation as an extension application for instance-level segmentation. Jia-Ching Wang 王家慶 2017 學位論文 ; thesis 63 zh-TW |
collection |
NDLTD |
language |
zh-TW |
format |
Others
|
sources |
NDLTD |
description |
碩士 === 國立中央大學 === 資訊工程學系 === 105 === In recent years, as a branch of machine learning, deep learning play an important role in Artificial Intelligence, which Convolutional Neural Network (CNN) has a breakthrough Performance in the image classification when comparing with traditional classification methods. The emergence of the full Convolutional Network (FCN)[10] also makes the study of image semantic segmentation flourish. In contrast to past work clustering according to the image texture and color, FCN joined the training of semantic information to improve the accuracy of semantic segmentation. Our paper combines the advantages of two networks, an object boundary based approach to strengthen the integrity of edge and the object itself, and the other is responsible for the prediction of image semantic segmentation, proposed an end-to-end training network architecture.
In this paper, proposed architecture improves the DT EdgeNet (Domain Transform with EdgeNet)[11]. Here, we combined the OBG-FCN [12] mask network and replaced the [11] edge network. The used mask network can predict background, object, and object edge reference diagrams. In addition, our architecture uses multi-scale ResNet-101 as the base network and introduces multi-scale Atrous Convolution to architecture training to preserve the dimensions of the feature map, which increases the receptive and further to enhance the accuracy of semantic segmentation.
In the experiments, we got the high performance of recognition on the VOC2012 test set. In addition, we combined extraction of object bounding box generated by Faster RCNN and result of proposed semantic segmentation as an extension application for instance-level segmentation.
|
author2 |
Jia-Ching Wang |
author_facet |
Jia-Ching Wang Kuan-Chung Wang 王冠中 |
author |
Kuan-Chung Wang 王冠中 |
spellingShingle |
Kuan-Chung Wang 王冠中 Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided |
author_sort |
Kuan-Chung Wang |
title |
Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided |
title_short |
Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided |
title_full |
Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided |
title_fullStr |
Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided |
title_full_unstemmed |
Semantic Segmentation Multi Scale Recurrent Convolutional Neural Network Based On Object Mask and Boundary Guided |
title_sort |
semantic segmentation multi scale recurrent convolutional neural network based on object mask and boundary guided |
publishDate |
2017 |
url |
http://ndltd.ncl.edu.tw/handle/4vcn3f |
work_keys_str_mv |
AT kuanchungwang semanticsegmentationmultiscalerecurrentconvolutionalneuralnetworkbasedonobjectmaskandboundaryguided AT wángguānzhōng semanticsegmentationmultiscalerecurrentconvolutionalneuralnetworkbasedonobjectmaskandboundaryguided AT kuanchungwang jīyúwùjiànzhēzhàoyǔbiānjièyǐndǎoduōchǐdùdìhuíjuǎnjīshénjīngwǎnglùzhīyǔyìfēngē AT wángguānzhōng jīyúwùjiànzhēzhàoyǔbiānjièyǐndǎoduōchǐdùdìhuíjuǎnjīshénjīngwǎnglùzhīyǔyìfēngē |
_version_ |
1719160899772612608 |