End-to-End Instance Segmentation with Transformers
Project Description
In this project, we propose an instance segmentation Transformer, which is the first end-to-end
framework for instance segmentation in histopathology images. Our results will be demonstrated on two public medical image datasets.
Supervisor
LI Xiaomeng
Quota
3
Course type
UROP1000
UROP1100
UROP2100
UROP3100
UROP4100
Applicant's Roles
The expectation of the applicant is to implement the algorithm with the help of a senior Ph.D. student. The applicant also can discuss ideas with senior Ph.D. students and advisors.
Applicant's Learning Objectives
1, The applicant is clear about deep learning and how to use it for medical image applications.
2, The applicant gets familiar with the next-generation transformer neural network.
3, Write a paper if possible.
Complexity of the project
Easy