Show simple item record

FieldValueLanguage
dc.contributor.authorDong, Minjing
dc.date.accessioned2019-09-11
dc.date.available2019-09-11
dc.date.issued2019-09-11
dc.identifier.urihttp://hdl.handle.net/2123/21089
dc.description.abstractHuman motion prediction is one of the key problems in computer vision and robotic vision and has received increasing attention in recent years. The target is to generate the future continuous, realistic human poses given a seed sequence, which can further assist human motion analysis. However, due to the high-uncertainty, it is difficult and challenging to model human dynamics which not only requires spatial information including complicated joint correlations, but also temporal information including periodic properties. Recently, deep recurrent neural networks (RNNs) have achieved impressive success in forecasting human motion with a sequence-to-sequence architecture. However, forecasting in longer time horizons often leads to implausible human poses or converges to mean poses, because of error accumulation and difficulties in keeping track of longer-term information. Based on these observations, in this study, we propose to retrospect human dynamics with attention. A retrospection module is designed upon RNN to regularly retrospect past frames and correct mistakes in time. This significantly improves the memory of RNN and provides sufficient information for the decoder networks to generate longer-term predictions. Moreover, we present a spatial attention module to explore cooperation among joints in performing a particular motion as well as a temporal attention module to exploit the level of importance among observed frames. Residual connections are also included to guarantee the performance of short-term prediction. We evaluate the proposed algorithm on the largest and most challenging Human 3.6M dataset in the field. Experimental results demonstrate the necessity of investigating motion prediction in a self-audit manner and the effectiveness of the proposed algorithm in both short-term and long-term predictions.en_AU
dc.publisherUniversity of Sydneyen_AU
dc.publisherFaculty of Engineeringen_AU
dc.publisherSchool of Computer Scienceen_AU
dc.rightsThe author retains copyright of this thesis. It may only be used for the purposes of research and study. It must not be used for any other purposes and may not be transmitted or shared with others without prior permission.en_AU
dc.subjectDeep learningen_AU
dc.subjectComputer Visionen_AU
dc.subjectHuman Motion Predictionen_AU
dc.subjectMocap Data Analysisen_AU
dc.subjectrecurrent neural networksen_AU
dc.titleModelling Skeleton-based Human Dynamics via Retrospectionen_AU
dc.typeMasters Thesisen_AU
dc.type.pubtypeMaster of Philosophy M.Philen_AU
dc.description.disclaimerAccess is restricted to staff and students of the University of Sydney . UniKey credentials are required. Non university access may be obtained by visiting the University of Sydney Library.en_AU


Show simple item record

Associated file/s

Associated collections

Show simple item record

There are no previous versions of the item available.