Deep learning methods with less supervision

To tackle the immense burden of acquiring accurate, pixel-level annotations for semantic segmentation tasks, we propose a weakly-supervised deep learning framework. We incorporate state-of-the-art foundational models to propagate pseudo-labels. Then, explore the viability of training a fully convolu...

Full description

Saved in:

Bibliographic Details
Main Author:	Chai, Youxiang
Other Authors:	Lin Guosheng
Format:	Final Year Project
Language:	English
Published:	Nanyang Technological University 2023
Subjects:	Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
Online Access:	https://hdl.handle.net/10356/172909
Tags:	Add Tag No Tags, Be the first to tag this record!
Institution:	Nanyang Technological University
Language:	English

id	sg-ntu-dr.10356-172909
record_format	dspace
spelling	sg-ntu-dr.10356-1729092024-01-05T15:37:11Z Deep learning methods with less supervision Chai, Youxiang Lin Guosheng School of Computer Science and Engineering gslin@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision To tackle the immense burden of acquiring accurate, pixel-level annotations for semantic segmentation tasks, we propose a weakly-supervised deep learning framework. We incorporate state-of-the-art foundational models to propagate pseudo-labels. Then, explore the viability of training a fully convolutional network based on our pseudo-labels. In addition, we experiment and evaluate the results of different loss functions and attempt the refinement of masks using conditional random fields. Bachelor of Engineering (Computer Science) 2023-12-31T07:54:34Z 2023-12-31T07:54:34Z 2023 Final Year Project (FYP) Chai, Y. (2023). Deep learning methods with less supervision. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/172909 https://hdl.handle.net/10356/172909 en SCSE22-0688 application/pdf Nanyang Technological University
institution	Nanyang Technological University
building	NTU Library
continent	Asia
country	Singapore Singapore
content_provider	NTU Library
collection	DR-NTU
language	English
topic	Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision
spellingShingle	Engineering::Computer science and engineering::Computing methodologies::Image processing and computer vision Chai, Youxiang Deep learning methods with less supervision
description	To tackle the immense burden of acquiring accurate, pixel-level annotations for semantic segmentation tasks, we propose a weakly-supervised deep learning framework. We incorporate state-of-the-art foundational models to propagate pseudo-labels. Then, explore the viability of training a fully convolutional network based on our pseudo-labels. In addition, we experiment and evaluate the results of different loss functions and attempt the refinement of masks using conditional random fields.
author2	Lin Guosheng
author_facet	Lin Guosheng Chai, Youxiang
format	Final Year Project
author	Chai, Youxiang
author_sort	Chai, Youxiang
title	Deep learning methods with less supervision
title_short	Deep learning methods with less supervision
title_full	Deep learning methods with less supervision
title_fullStr	Deep learning methods with less supervision
title_full_unstemmed	Deep learning methods with less supervision
title_sort	deep learning methods with less supervision
publisher	Nanyang Technological University
publishDate	2023
url	https://hdl.handle.net/10356/172909
_version_	1787590726269272064

Deep learning methods with less supervision

Similar Items