Applications of AI for image processing

Course guarantor:

Petr Kubera

Lecturer:

Petr Kubera

Tutor:

Petr Kubera

Supervisor:

Department of Software Engineering

Synopsis:

The aim of the course is to provide students with hands-on experience with existing state-of-the-art tools based on neural networks (NN), with a primary focus on image processing. An example of such tools in the NVIDIA ecosystem is the Jetson device. Students will learn how to set up this device, install the required software, and develop programs that use existing pre-trained models.

Requirements:

Credit is awarded for attendance (at least 80%, a necessary but not sufficient condition) and for developing an application. The topic of the application is specified during the course.

Syllabus of lectures:

1. Introduction to the device (approx. 1 day)

a) Setup and OS installation

b) Basic operation introduction to the Linux console, basic commands

c) Software installation using Docker containers, basics of Docker usage

2. Introduction to basic pre-installed NN models (approx. 1 day)

a) General device architecture and description of technologies

b) Running models and adjusting their parameters

c) Installing additional models

d) Using the programming API to access information

3. Developing simple applications using NN models (1 day)

a) More detailed introduction to the APIs of individual models

b) Creating simple applications using object detection, image segmentation, etc., based on NVIDIA tutorials

4. Independent development of a more complex program (2 days)

Students independently develop a more complex application

Syllabus of tutorials:

Study Objective:

Study materials:

Note:

Time-table for winter semester 2025/2026:

Time-table is not available yet

Time-table for summer semester 2025/2026:

Time-table is not available yet

The course is a part of the following study plans:

• Aplikace informatiky v přírodních vědách (elective course)