Deep Learning for Computer Vision

Monday 6 - 7:45 pm // Use deep learning approaches on a computer vision application, learn the basics and create your own project.


Für diesen Kurs kann man keine Teilnahmenbescheinigung erhalten.

Bei ausreichender Teilnahme und Erbringung eines der angebotenen Leistungsnachweise kannst Du ein Leistungszertifikat erwerben.



Our courses are open to all and free of charge for students, pupils, trainees, founders, scientific staff, job seekers, refugees, and pensioners. For employees and self-employed persons, a fee of 200 euros is charged when acquiring a certificate of achievement. Employees and self-employed persons can apply for a scholarship via

What you get

This course extends the fourth course (Convolutional Neural Networks of the online Coursera Specialization "Deep Learning" provided by

The course is a mix of Coursera content, additional materials prepared for you and your own project. We will start by learning the basics of a Convolutional Neural Network (CNN), the reason behind their diffusion, and having an overview of the classical architectures. Then we discuss about how to train a CNN, their advantages and limitations, and different approaches to this topic. A large part of the course will be devoted to applications for computer vision, we will see examples of object detection, face recognition, image segmentation and neural style transfer.

You will be supported in the creation of a group project during the course. We will have discussion round, peer review checkpoints and some hints for the deploying your project.

This course is part of the Machine Learning Degree. Participants of the program for the Machine Learning Degree get preferred access to this course. Find more information on the Machine Learning Degree page. For example, the page for this course is this one and each week has a subpage with the material and useful links.

All needed software and online course content is free. As Germany's first Coursera for Campus partner, will in particular provide you with full access to the Coursera online learning platform.

What you should bring

This course is the natural follow-up of the the first course Deep Learning from Scratch.

If you successfully finished the first course Deep Learning from Scratch, you have nothing to worry about and your application will be granted (as long as there are available places).

If you did not participate in the first course Deep Learning from Scratch, you are welcome to apply, just note that it will be assumed that you have some experience with neural networks (not with CNN, but at least with general machine learning applications).

During the course you will be expected to work through the assigned online course content and complete home assignments on a weekly basis, plus you will work on a group project (you can decide how to distribute the time yourself). For keeping up with the course we usually recommend to plan 5 to 10 hours each week, depending on your preparation and background.

In order to receive a certificate of achievement ("Leistungszertifikat"/ ECTS) for this course, active participation is expected, no more than two classes may be missed, and you have to conduct a practice project in a team of 2 to 4 persons. At the end of the course the project has to be presented and a well documented project source code has to be submitted.

Before the first meeting there will be a Semester Kick-Off Event on October 21.

In the Kick-Off you will meet your course instructor, can ask questions about the participation in the course, and get helpful information to prepare for the start of the course. The attendance at the Kick-Off is not mandatory but recommended for all participants. During the current Coronavirus outbreak, the Kick-Off as well as the course will be streamed online via Zoom. Consultation hours will be organized to allow also personal meeting and an easier communication.


Deep Learning for Computer Vision
Wintersemester 2021/2022
18:00 - 19:45 Uhr


Luca Palmieri
Start: 25.10.2021
Ende: 17.01.2022
**Content:** Basics of CNN: convolution, edge detection, strides, pooling, layers **Homework:** Putting together different layers to create the first example of a CNN architecture

Luca Palmieri

**Content:** Additional material on convolutional models **Homework:** Training two convolutional models to recognize smiles and hands images which shows numbers

Luca Palmieri

**Content:** Classical CNN Architectures like ResNet, InceptionNetwork, MobileNet and EfficientNet **Homework:** Training with a Residual Network

Luca Palmieri

**Content:** Additional discussion on pros & cons of transfer learning **Homework:** Transfer Learning with MobileNet

Luca Palmieri

**Content:** Detection, localization, bounding box, intersection, union, discussion of various concepts for detection algorithms **Homework:** Car detection using YOLO implementation

Luca Palmieri

**Content:** Checkpoint on the project, peer review of the progress **Homework:** Image segmentation using U-Net

Luca Palmieri

**Content:** Face recognition, one shot learning, siamese networks and triplet loss **Homework:** Face recognition using CNN

Luca Palmieri

**Content: ** Neural style transfer, content and style cost functions **Homework:** Art Generation with Neural Style Transfer

Luca Palmieri

**Content:** Final presentation of the projects of each group **Homework:** Finishing the project and preparing the presentation

Luca Palmieri