Tools
Tools
Bauskar, Dhanshree (2022) Dance Video Classification into Relevant Street Dancing Styles using Deep Learning Techniques. Masters thesis, Dublin, National College of Ireland.
Preview |
PDF (Master of Science)
Download (1MB) | Preview |
Preview |
PDF (Configuration manual)
Download (720kB) | Preview |
Abstract
Computer vision is the emerging research area in technical field, and video classification is considered as one of the categories of computer vision. In past few years, deep learning approaches have elongated in video domain. Combination of deep learning approaches and computer vision have solved complex tasks. Consequently, deep learning approach is used to solve computer vision problem like video classification. As a increased people’s interest in the field of dance, this research intends to classify 10 different dance styles using the database provided by AIST dance academy. Dance as a whole consists of Body movements, musical pieces, facial expressions and hand gestures which makes it complex problem to classify dance videos. Using VGG-16 model for pre-processing and training of data gave the accuracy result of 75.86% followed by the accuracy obtained by VGG-19 model that is 68.96% comparatively Convolutional neural network has the lowest accuracy. The model is evaluated with different evaluation criteria.
| Item Type: | Thesis (Masters) |
|---|---|
| Subjects: | G Geography. Anthropology. Recreation > GV Recreation Leisure M Music and Books on Music > M Music Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning |
| Divisions: | School of Computing > Master of Science in Data Analytics |
| Depositing User: | Tamara Malone |
| Date Deposited: | 18 Jan 2023 16:28 |
| Last Modified: | 06 Mar 2023 16:58 |
| URI: | https://norma.ncirl.ie/id/eprint/6083 |
Actions (login required)
 |
View Item |