|
Tracking Olympiad (TRACO)5 ECTS (englische Bezeichnung: Tracking Olympiad)
Modulverantwortliche/r: Andreas Kist Lehrende:
Andreas Kist, René Groh
Startsemester: |
SS 2022 | Dauer: |
1 Semester | Turnus: |
jährlich (SS) |
Präsenzzeit: |
60 Std. | Eigenstudium: |
90 Std. | Sprache: |
Englisch |
Lehrveranstaltungen:
-
-
Tracking Olympiad
(Seminar, 4 SWS, Andreas Kist et al., Di, 10:15 - 11:45, Raum n.V.; Fr, 9:15 - 10:45, Raum n.V.; AIBE Seminar Room, Werner-von-Siemens-Str. 61, 91054 Erlangen)
Empfohlene Voraussetzungen:
Es wird empfohlen, folgende Module zu absolvieren, bevor dieses Modul belegt wird:
Data Science Survival Skills (WS 2021/2022)
Deep Learning (WS 2021/2022)
Pattern Recognition (WS 2021/2022)
Introduction to Machine Learning (WS 2021/2022)
Machine Learning for Engineers I: Introduction to Methods and Tools (WS 2021/2022)
Machine Learning for Engineers II: Advanced Methods (WS 2021/2022)
Inhalt:
Computer vision is one of the major tasks and applications of artificial intelligence (AI). Gaining hands-on experience is therefore of great importance for future AI developers. In the Tracking Olympiad, students utilize latest object detection and tracking algorithms to track a freely, randomly moving object (“HexBug”) in a given arena. The students will be provided with a set of videos that contain the ground-truth positional information and implement an own tracking technique.
At the beginning of the seminar, all students are divided into teams which compete with each other to find the best strategy for tracking the HexBug. The team’s tracking prediction needs to be an algorithm that incorporates each student’s tracking algorithm. The team’s score will be evaluated by applying the team’s tracking algorithm to previously unseen/withheld videos. Further, the team acquires and annotates own data to improve their tracking algorithms. Each team selects videos that are tested by the other teams’ algorithm and are subsequently ranked similar to a soccer league table. The aim of this seminar is to enable each student developing an own AI-powered tracking algorithm that is an integral part of a team solution.
The Tracking Olympiad consists of two sessions in a given week, one with a journal club explaining AI tracking concepts by students and one for open Q&A depending on the individual student’s progress with voluntary developmental time.
Lernziele und Kompetenzen:
Students
will be able to create own code
are able to create acquire and annotate own data
can document their code
will strengthen their team skills
can develop tracking algorithms
will learn about latest AI methods
can present complex topics
can extract relevant information from journal papers
Literatur:
- Burger and Burge, Principles of Digital Image Processing (all volumes)
Howes and Minichino, Learning OpenCV 4 Computer Vision with Python 3
Sebastian Raschka, Python Machine Learning: Machine Learning and Deep Learning with Python, scikit-learn, and TensorFlow 2
Aurélien Géron, Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow
Pereira et al., Quantifying behaviour to understand the brain, Nat Neurosci 2020
Verwendbarkeit des Moduls / Einpassung in den Musterstudienplan: Das Modul ist im Kontext der folgenden Studienfächer/Vertiefungsrichtungen verwendbar:
- Artificial Intelligence (Master of Science)
(Po-Vers. 2021s | TechFak | Artificial Intelligence (Master of Science) | Gesamtkonto | Hauptseminar | Tracking Olympiad)
- Data Science (Master of Science)
(Po-Vers. 2021w | Gesamtkonto | Anwendungsfächer | Artificial intelligence in biomedical engineering (AIBE) | Tracking Olympiad)
- Medizintechnik (Master of Science)
(Po-Vers. 2018w | TechFak | Medizintechnik (Master of Science) | M4 Hauptseminar Medizintechnik | Tracking Olympiad)
- Medizintechnik (Master of Science)
(Po-Vers. 2019w | TechFak | Medizintechnik (Master of Science) | Modulgruppe M4 - Hauptseminar | Hauptseminar Medizintechnik / Advanced Seminar Medical Engineering | Tracking Olympiad)
Studien-/Prüfungsleistungen:
Tracking Olympiad (Prüfungsnummer: 76121)
(englischer Titel: Talk and written report)
- Prüfungsleistung, Seminarleistung, Dauer (in Minuten): 20, benotet, 5 ECTS
- Anteil an der Berechnung der Modulnote: 100.0 %
- weitere Erläuterungen:
Talk (presenting paper/video) 20 min, written report 10-15 pages, valued 50% talk and 50% written report for grading
- Prüfungssprache: Englisch
- Erstablegung: SS 2022, 1. Wdh.: SS 2023
|
|
|
|
UnivIS ist ein Produkt der Config eG, Buckenhof |
|
|