Evaluation of Human Pose Estimation Methods in Video Data – Bc. Zuzana Tunegová
Bc. Zuzana Tunegová
Master's thesis
Evaluation of Human Pose Estimation Methods in Video Data
Evaluation of Human Pose Estimation Methods in Video Data
Abstract:
Cieľom tejto práce je zhodnotiť kvalitu existujúcich metód na detekciu ľudských postáv na obrázkoch obsahujúcich viacero jednotlivcov. Vyberáme päť sľubných metód a overujeme ich kvalitu pomocou metrík average precision a average recall. Pri vyhodnocovaní metrík sú použité dva datasety: COCO a CrowdPose. Implementujeme funkciu, ktorá vypočíta average precision a average recall pomocou podobnosti kľúčových …moreAbstract:
This thesis aims to evaluate the quality of existing pose estimation methods containing multiple individuals. We select five promising methods and validate them using average precision and average recall metrics. The metrics are evaluated on two benchmarking datasets, COCO and CrowdPose. We implement a function that calculates the average precision and average recall using object keypoint similarity …more
Language used: English
Date on which the thesis was submitted / produced: 15. 12. 2023
Identifier:
https://is.muni.cz/th/jz6l2/
Thesis defence
- Date of defence: 12. 2. 2024
- Supervisor: doc. RNDr. Jan Sedmidubský, Ph.D.
- Reader: doc. RNDr. Vlastislav Dohnal, Ph.D.
Citation record
Full text of thesis
Contents of on-line thesis archive
Published in Theses:- světu
Other ways of accessing the text
Institution archiving the thesis and making it accessible: Masarykova univerzita, Fakulta informatikyMasaryk University
Faculty of InformaticsMaster programme / field:
Software Engineering / Design and development of software systems
Theses on a related topic
- No theses on a related topic available.