Entry requirements

120 credits. Participation in Deep Learning and Neural Networks. Participation in on of the courses Introduction to Image Analysis and Computer-Assisted Image Analysis. Proficiency in English equivalent to the Swedish upper secondary course English 6.

Learning outcomes

After passing the course the student should be able to:

• use modern environments for deep machine learning to solve practical image processing and image analysis problems; 
• describe and use deep convolutional networks for image classification, object detection and image segmentation; describe and use different kinds of regularisation techniques for visual data; 
• describe and use methods for image generation; 
• explain and use methods for interpretation and understanding of deep models; 
• critically analyse research in deep learning for image processing, assess its possibilities and limitations. 

Content

Deep learning for visual data. Convolutional neural networks (CNN) and methods for training them, transfer learning and data augmentation for visual data. Different architectures and applications in image analysis (classification, detection, segmentation). Deep generative models. Visualisation and understanding of deep neural networks. Possibilities and limitations with deep learning.

Instruction

Lectures, assignments, computer exercises and project work in groups.

Assessment

Written exam, assignments, oral and written presentation of project work.

Transitional provisions

Cannot be included in the same degree as 1MD120 Deep Learning for Image Analysis.