Beyond Supervised Learning: Data-Driven Techniques for Image Segmentation with Fewer Labels
March 1st 2021 (9:45 AM - 11:25 AM CST Time)
March 1st 2021 (9:45 AM - 11:25 AM CST Time)
OVERVIEW
OVERVIEW
In the last few years, machine learning has experienced an astonishing development in all domains. It has become the technique to-go for most of existing image analysis problems reporting state-of-the-art results in tasks such as classification, detection and segmentation. The superior performance is reported in supervised learning, however, this paradigm relies on the strong assumption of having a well-representative annotated dataset. In several real world problems such as image segmentation this assumption is not practical, and the bias in the annotations adversely affect the segmentation outputs. We then motivate the use of fewer labels for segmenting several types of imaging data. In particular, we study the advances of recent unsupervised and semi-supervised methods that exploit large sets of unlabelled data for segmentation. We will discuss the current state-of-the-art techniques, challenges and opportunities.
In the last few years, machine learning has experienced an astonishing development in all domains. It has become the technique to-go for most of existing image analysis problems reporting state-of-the-art results in tasks such as classification, detection and segmentation. The superior performance is reported in supervised learning, however, this paradigm relies on the strong assumption of having a well-representative annotated dataset. In several real world problems such as image segmentation this assumption is not practical, and the bias in the annotations adversely affect the segmentation outputs. We then motivate the use of fewer labels for segmenting several types of imaging data. In particular, we study the advances of recent unsupervised and semi-supervised methods that exploit large sets of unlabelled data for segmentation. We will discuss the current state-of-the-art techniques, challenges and opportunities.
ORGANISING TEAM
ORGANISING TEAM
LINE UP SPEAKERS
LINE UP SPEAKERS
SCHEDULE
SCHEDULE
Talk 1: Learning from Sparse Annotations for Semantic Segmentation of Remote Sensing Images
Talk 1: Learning from Sparse Annotations for Semantic Segmentation of Remote Sensing Images
9:45AM-10:05AM
Yuansheng Hua, German Aerospace Center (DLR), Germany
Talk 2: A Semisupervised Approach for Video Semantic Segmentation
Talk 2: A Semisupervised Approach for Video Semantic Segmentation
10:05AM-10:25AM
Rihuan Ke, University of Cambridge, United Kingdom
Talk 3: Semi-Supervised Semantic Segmentation Needs Strong, Varied Perturbations
Talk 3: Semi-Supervised Semantic Segmentation Needs Strong, Varied Perturbations
10:25AM-10:45AM
Goffrey French, University of East Anglia, United Kingdom
Talk 4: Learning Across Domains for Semantic Segmentation in Urban Roads
Talk 4: Learning Across Domains for Semantic Segmentation in Urban Roads
10:45AM-11:05AM
Tarun Kalluri, University of California, San Diego, U.S.
REGISTRATION
REGISTRATION
Registration is required. If you want to register to the SIAM-CSE21 you can do it in: LINK SIAM-CSE21
Registration is required. If you want to register to the SIAM-CSE21 you can do it in: LINK SIAM-CSE21