Full metadata record
DC FieldValueLanguage
dc.contributorDepartment of Electronic and Information Engineeringen_US
dc.contributor.advisorChi, Zheru (EIE)en_US
dc.creatorHe, Jiacheng-
dc.publisherHong Kong Polytechnic Universityen_US
dc.rightsAll rights reserveden_US
dc.titleDeep learning in basketball action recognitionen_US
dcterms.abstractBasketball is a sport that many people love, and I also like it very much. There are many different actions in this sport, such as shooting, dribbling, passing, etc. In daily competitions, these data are calculated manually. If the deep learning model can be applied to basketball action recognition, it can replace some manual data processing and data entries, make it more intelligent, and bring convenience to the audience to watch the game. And the technique can be extended to other sports with a good prospect. 3D CNNs are very powerful at extracting features in videos and applying them. It is suitable for the action recognition of basketball mentioned above. Convolutional neural networks with spatio-temporal 3D kernels (3D CNNs) are used to extract spatio-temporal features in the video to recognize the actions of sport men/women in the video. The model trained by the 3D kernel is prone to overfitting due to its large number of parameters. Therefore, a model combining residual networks (ResNets) with 3D CNNs, which has excellent performance in handling over-fitting problems in 2D convolutional learning, achieve a relatively better performance. This dissertation introduces the use of 3D ResNets to train SpaceJam, which is a dataset for basketball action recognition and show its training and test results.en_US
dcterms.extentiv, 44 pages : color illustrationsen_US
dcterms.isPartOfPolyU Electronic Thesesen_US
dcterms.educationalLevelAll Masteren_US
dcterms.LCSHHuman activity recognitionen_US
dcterms.LCSHPattern recognition systemsen_US
dcterms.LCSHImage processing -- Digital techniquesen_US
dcterms.LCSHHong Kong Polytechnic University -- Dissertationsen_US
dcterms.accessRightsrestricted accessen_US

Files in This Item:
File Description SizeFormat 
5659.pdfFor All Users (off-campus access for PolyU Staff & Students only)1.2 MBAdobe PDFView/Open

Copyright Undertaking

As a bona fide Library user, I declare that:

  1. I will abide by the rules and legal ordinances governing copyright regarding the use of the Database.
  2. I will use the Database for the purpose of my research or private study only and not for circulation or further reproduction or any other purpose.
  3. I agree to indemnify and hold the University harmless from and against any loss, damage, cost, liability or expenses arising from copyright infringement or unauthorized usage.

By downloading any item(s) listed above, you acknowledge that you have read and understood the copyright undertaking as stated above, and agree to be bound by all of its terms.

Show simple item record

Please use this identifier to cite or link to this item: https://theses.lib.polyu.edu.hk/handle/200/11183