M2. Optimization and Inference techniques in Computer Vision (31819)

The aim of this module is to learn about numerical optimization algorithms that are behind many tasks in computer vision. Main concepts will include energy minimization, numerical techniques for variational problems, convex optimization and graphical models. These techniques will be applied in the project in the context of image segmentation and restoration (denoising and inpainting).
Project title: 
Removing objects in urban scenes
The goal of this project is to gain practical experience with the basic optimization methods used in computer vision. The methods will be studied in the context of image segmentation and restoration (denoising and inpainting), with special emphasis on the formulation of the optimization problem and its resolution. The tools learnt along this project are generic and present in a majority of computer vision applications (as found in other modules, e.g. in clutter removal for improved 3D scene reconstruction, M6).
Module lectures: 

 

Academic Year 2017-2018        
Week Date Time Lecture Lecturer University Building Room
1 Tue. Oct. 3rd     

 

     
Thu. Oct. 5th 16:00 -18:00 Introduction to energy minimization methods. Overview of varational formulation. Numerical techniques for variational problems through examples (I); gradient descent Juan Fco. Garamendi UPF Roc Boronat 52221
Thu. Oct. 5th 18:00 - 19:00

Project Introduction

Karim Lekadir

UPF Roc Boronat

52.201

52.213

               
2 Tue. Oct. 10th 16:00 -18:00 Numerical techniques for variational problems through examples (II); Gateaûx derivate, Euler-Lagrange equation, multigrid. Applications (ROF denoising, image inpainting and Poisson editing) Juan Fco. Garamendi UPF Roc Boronat  52119
Thu. Oct. 12th 16:00 -18:00 HOLIDAY        
               
 3 Tue. Oct. 17th  16:00 -18:00

Review of numerical linear algebra: minimum squares, regression, singular, value decomposition, interative methods,applications. 

Juan Fco. Garamendi  UPF Roc Boronat  52325
Thu. Oct. 19th  16:00 -18:00 Convex optimization (I). Constrained and unconstrained optimization. Primal, dual, and primal dual methods. Applications: Total Variation restoration, disparity computation, optical flow computation,... Coloma Ballester UPF  Roc Boronat  52119
Thu. Oct. 19th 18:00 - 19:00 Project follow-up Karim Lekadir      52.117
               
4 Tue. Oct. 24th 16:00 -18:00 Convex optimization (II): Primal, dual, and primal dual methods. Convex relaxation. Segmentation with variational models. The Mumford and Shah Functional and the Level sets framework.  Coloma Ballester UPF  Roc Boronat  52325
Thu. Oct. 26th 16:00 - 18:00 Bayesian networks and MRFs. Inference Types. Example: stereo, denoising.  Joan Serrat UPF  Roc Boronat  52019
Thu. Oct. 26th 18:00 - 19:00 Project follow-up

Karim Lekadir

UPF  Roc Boronat 52.117
               


5

Tue. Oct. 31st 16:00 -18:00 Inference algorithms (I): belief propagation (sum-product and max-sum) and generalizations. Examples: blob tracking, video synchronization. Guidelines for PGM exercises.  Oriol Ramos      52325
Thu. Nov. 2nd 16:00 -18:00 Inference algorithms (II): Graph cuts, linear programming relaxation, Monte Carlo methods. Examples: co-segmentation Oriol Ramos UPF  Roc Boronat  52119
Thu. Nov. 2nd  18:00 - 19:00 Project follow-up

Karim Lekadir

UPF  Roc Boronat

52.117

               
6 Tue. Nov. 7th  16:00-18:00 Learning of graphical models.Structured SVMs Joan Serrat UPF Roc Boronat  52325
Thu. Nov. 9th 16:00-18:00 PGM exercises: inference for segmentation and learning for labeling

Oriol Ramos 

Joan Serrat

UPF Roc Boronat

54005

 

Thu. Nov. 9th 18:00 -19:00 Project follow-up

Karim Lekadir

UPF Roc Boronat

 52.119

               
7 Thu. Nov.16th 16:00-19:00 Project Presentations 

Karim Lekadir

UPF Roc Boronat

52119

               
8 Tue. Nov. 21st  16:00-19:00 HOMEWORK        
Thu. Nov. 23rd  16:00-19:00 HOMEWORK        
               
9 Thu. Nov. 30th 16:00 -19:00 EXAM Coloma Ballester UPF  Roc Boronat  52023
               

 M2 Student Guide [here]