High-Level Computer Vision

 

Overview

This course will cover essential techniques for high-level computer vision. These techniques facilitate semantic interpretation of visual data, as it is required for a broad range of applications like robotics, driver assistance, multi-media retrieval, surveillance etc. In this area, the recognition and detection of objects, activities and visual categories have seen dramatic progress over the last decade. We will discuss the methods that have lead to state-of-the-art performance in this area and provide the opportunity to gather hands-on experience with these techniques.

Course Information

Semester:  SS

Year:  2019

Lecture start:  Wednesday April 10

Tutorial start:  Monday April 15

 

Time and Location:

lecture: Wednesdays 10:00 - 12:00 E 1.4 room 024 (start at 10:15)

tutorial: Mondays 10:00 - 12:00 E 1.4 room 024

On May 29, the lecture will be in E 1.4 room 022.

 

Mailing List: send an email with your matriculation number and full name to rshetty@mpi-inf.mpg.de with [hlcv-subscribe] in the subject.

Exam: July 18 / 19, August 20 / 21, October 1 / 2

Registration: send an email to  rshetty@mpi-inf.mpg.de  with [hlcv exam registration] in the subject

 

Lecturer(s):   Prof. Dr. Bernt Schiele and Dr. Mario Fritz

TA(s):              Rakshith Shetty (office hour: Thursdays 15:00-16:00 E 1.4 room 628)

                         Yang He (office hour: Wednesdays 15:00-16:00 E 1.4 room 608)

 

Lectures

  • 2019-04-10: Introduction (slides)

  • 2019-04-17: Image Classification, Linear Classifer, Losses (slides v1.1 with corrections)

  • 2019-04-24: Backpropagation, Deep Learning Intro, CNNs (slides)

  • 2019-05-08: Convolutional Neural Networks, Network Visualization, Feature Generalization (slides)

  • 2019-05-15: Object Detection and Semantic Segmentation (slides)

  • 2019-05-22: CNN Architectures (slides)

Exercises