Multi-class image segmentation in EM images
Event details
| Date | 22.05.2018 |
| Hour | 11:00 › 13:00 |
| Speaker | Pamuditha Udaranga Wickramasinghe |
| Location | |
| Category | Conferences - Seminars |
EDIC candidacy exam
Exam president: Prof. Wulfram Gerstner
Thesis advisor: Prof. Pascal Fua
Co-examiner: Dr. Graham Knott
Abstract
We study the problem of multi-class image segmentation in Electron Microscopy(EM) images with the focus on extracting intricate structures made up of multiple parts. The problem is formulated as a structured prediction task and we review two approaches to solve it, namely (1) Probabilistic Graphical Models and (2) Deformable Models. Furthermore, we propose a framework for learning structural relationships from data. We present the results based on this preliminary work and discuss future directions for our research.
Background papers
Deformable models in medical image analysis: a survey, Medical Image Analysis 1996, by T. McInerney, D. Terzopoulos
Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials, NIPS 2011, by P. Krähenbühl, V. Koltun.
Segmentation-Aware Convolutional Networks Using Local Attention Masks, ICCV 2017, by A. W. Harley, K. G. Derpanis, I. Kokkinos
Exam president: Prof. Wulfram Gerstner
Thesis advisor: Prof. Pascal Fua
Co-examiner: Dr. Graham Knott
Abstract
We study the problem of multi-class image segmentation in Electron Microscopy(EM) images with the focus on extracting intricate structures made up of multiple parts. The problem is formulated as a structured prediction task and we review two approaches to solve it, namely (1) Probabilistic Graphical Models and (2) Deformable Models. Furthermore, we propose a framework for learning structural relationships from data. We present the results based on this preliminary work and discuss future directions for our research.
Background papers
Deformable models in medical image analysis: a survey, Medical Image Analysis 1996, by T. McInerney, D. Terzopoulos
Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials, NIPS 2011, by P. Krähenbühl, V. Koltun.
Segmentation-Aware Convolutional Networks Using Local Attention Masks, ICCV 2017, by A. W. Harley, K. G. Derpanis, I. Kokkinos
Practical information
- General public
- Free
Contact
- EDIC - [email protected]