Institutional Repository of Vilnius University
Žmogaus judėjimo generavimas naudojant neuroninius tinklus /
Abstract
dc:descriptionHuman motion synthesis is an important area of research for both fields of computer graphics and robotics as well as computer modeling. This work includes a comparative analysis of algorithms for human motion synthesis: the main principles of these methods are outlined as well as their strengths and weaknesses. 2 algorithms with the best desired quantities are identified. These methods are based on convolutional and recurrent neural networks. In order to compare them quantitatively, a new metric is proposed, the results of which are then compared to a conducted user survey. Finally, some improvements on configuration on the best performing recurrent network are proposed.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2020
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Krivickas, Vytautas,
- Contributors dc:contributor
-
- Būtėnas, Linas
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD81705453&prefLang=en_US