Institutional-Repository, University of Moratuwa.  

Action recognition by single stream convolutional neural networks : an approach using combined motion and static information

Show simple item record

dc.contributor.author Ramasinghe, S
dc.contributor.author Rodrigo, R
dc.date.accessioned 2018-11-07T21:01:06Z
dc.date.available 2018-11-07T21:01:06Z
dc.identifier.uri http://dl.lib.mrt.ac.lk/handle/123/13655
dc.description.abstract We investigate the problem of automatic action recognition and classification of videos. In this paper, we present a convolutional neural network architecture, which takes both motion and static information as inputs in a single stream. We show that the network is able to treat motion and static information as different feature maps and extract features off them, although stacked together. We trained and tested our network on Youtube dataset. Our network is able to surpass state-of-the-art hand-engineered feature methods. Furthermore, we also studied and compared the effect of providing static information to the network, in the task of action recognition. Our results justify the use of optic flows as the raw information of motion and also show the importance of static information, in the context of action recognition. en_US
dc.language.iso en en_US
dc.title Action recognition by single stream convolutional neural networks : an approach using combined motion and static information en_US
dc.type Conference-Abstract en_US
dc.identifier.faculty Engineering en_US
dc.identifier.department Department of Electronic and Telecommunication Engineering en_US
dc.identifier.year 2015 en_US
dc.identifier.conference 3rd IAPR Asian Conference on Pattern Recognition - 2015 en_US
dc.identifier.place Kuala Lumpur en_US
dc.identifier.pgnos pp. 101 - 105 en_US
dc.identifier.email samramasinghe@gmail.com en_US
dc.identifier.email ranga@uom.lk en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record