CSL462_618_2017

August-December 2017

CSL 462 - Computer Vision (UG)

CSL 618 - Computer Vision (PG)

Lectures (NIELIT Campus)

Wednesday – H8 – 2:55-3:45PM

Thursday – H8 – 3:50-4:40PM

Friday – H8 – 4:45-5:35PM

Labs

Thursday 09:00-10:45 AM - Lab 2

Friday 10:50-12:35 AM - Lab 3

Teaching Assistants

Amanjot Kaur (amanjot.kaur at iitrpr)

Shreya Ghosh (shreya.ghosh at iitrpr)

Showcase Assignment Submissions

Assignment 1 - Naman Goyal Sujit Rai Devendra Pratap Yadav

Assignment 2 - Shivam Mittal Prashant Patil Jaspinder Kaur

Assignment 3 - Akshay Dudhane Mayank Kumar Rajat Sharma

Slides (Can be accessed using IITRPR account only)

Mid-term topics survey (29/09/2017) [LINK]

End-term summary (27/11/2017) [LINK]

Reading material above point to Richard Szeliski's book until unless explicitly specified.

Marks distribution

Note the following can change on the instructor's desecration.

CSL 462

Assignments – 20%

Mid-term – 20%

End-term – 20%

Project – 40%

CSL 618

Assignments – 30%

Mid-term – 10%

End-term – 20%

Project – 40%

5% extra marks for research paper quality work in the project/assignment!

CSL 462 and 618 will have 2 different assignments

Assignment code in Matlab

Project can be in Matlab, C/C++ or Android/iOS

Minimum pass marks – 40/100

Minimum attendance – 70% (institute requirement)

Plagiarism

First instance of assignment plagiarism will lead to marks deduction.

Second instance of assignment plagiarism will lead to F.

Project plagiarism will be awarded F.

An interesting and quick read on how not to plagiarize - http://advice.writing.utoronto.ca/using-sources/how-not-to-plagiarize/

Assignment format – All reports have to be created in Latex by following the BMVC extended abstract.

Project can be both individual and group (2/3 max). List of projects will be shared (22/08/2017).

You can propose a new idea!

Project proposal to be submitted by 29/08/2017

The marking criteria will be shared with the project ideas.

Books

Richard Szeliski, Computer Vision: Algorithms and Applications, Springer 2010

Available free online: szeliski.org/book

Reference book:

Computer Vision: Modern Approach, Jean and Ponce

Other resources

Linear algebra refresher - https://stanford.edu/~arbenson/refresher/la-refresher-slides.pdf

Topic to be covered

Note the following can change on the instructor's desecration.

Introduction to Computer Vision, history and current

Image filtering and matching

Edge detection and pyramids

Keypoint detection and scale invariant keypoints

Local descriptors, robust matching, homographies

Camera models and stereo

Machine learning from computer vision perspective: introduction

Image recognition and Bag of words

Face detection and analytics

Pose detection and attributes

Computer Vision and affect

Deep learning in computer vision

Applications

Research papers

Acknowledgement - "Stand on the shoulders of the giants"

This course is based on materials from Prof. Mubarak Shah, Prof. James Hays, Prof. Roland Goecke, Prof. Steve Seitz, Prof. Kristen Grauman and Prof. Sanja Fidler's amazing computer vision courses.