Self-Supervised Learning for Semantic Segmentation of Images
Date
2023-09-22
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
0009-0009-4459-5257
Type
Thesis
Degree Level
Masters
Abstract
Artificial Neural Networks (ANN) are powerful Machine Learning (ML) models that can help solve problems that are hard or even impossible to design solutions for by hand. These models learn to exploit
information present in their target datasets to solve various problems. However, labelling data can be quite
expensive, time-consuming and often requires a domain expert. Therefore it would be quite beneficial if one
could train a model in such a way that exploits unlabeled data. Fortunately, Self-Supervised Learning (SSL)
methods are a family of learning algorithms that attempt to do just that. Many SSL methods exist, but in
this thesis, we explore Barlow Twins (BT) — a siamese network based on redundancy reduction, and Image
Reconstruction (IR) — a method proposed in Karnam’s thesis. In addition, we extend the Image Reconstruction method with both Coarse Cutout and Hide-and-Seek augmentations as they have been applied in similar
supervised and weakly-supervised segmentation task scenarios. We apply these methods and investigate the
results with the PASCAL VOC dataset.
Description
Keywords
Neural Networks, Machine Learning, Self-Supervised Learning, Semantic Segmentation
Citation
Degree
Master of Science (M.Sc.)
Department
Computer Science
Program
Computer Science