M2. Optimization and Inference for Computer Vision

The aim of this module is to learn about the optimization algorithms and inference techniques that are behind many tasks in computer vision. The main concepts will include energy formulation and minimization, numerical techniques for variational problems, gradient descent optimization algorithms and tools useful for deep learning strategies. convex optimization, and graphical models. These techniques will be applied in the project in the context of image segmentation and inpainting.

Module Project: Removing Objects in Natural and Urban Scenes

The aim of the project is to put into practice some of the lessons learned on optimisation and inference in the field of computer vision. Specifically, various optimisation techniques will be studied and implemented to solve computer vision problems such as image inpainting, Poisson editing and image segmentation. The students will learn how to test and optimise these techniques by using a wide range of natural images and real-life scenarios with varying characteristics and levels of complexity. They will also learn to analyse the implemented techniques and to understand their benefits and limitations in the age of big data.

M2 Schedule – Academic Year 2021-2022 – Student Guide <here>