New Zealand Journal of Agricultural Research abstracts
3D motion analysis based on machine learning in agriculture
informatisation
Jian Xiang
ZheJiang University of Science and Technology
310023, Hangzhou
PR China
and
ZheJiang University
310027, Hangzhou
PR China
freexiang@gmail.com
Abstract In this paper, a motion retrieval system is
investigated from a multiple-instance learning view. To retrieve
similar motion data, each human joint’s motion clip is regarded as a
bag, while each of its segments is regarded as an instance. First 3D
temporal-spatial features and their keyspaces of each human joint are
extracted. Then, data driven decision trees based on ensemble multiple
instance are automatically constructed to reflect the influence of each
point during the comparison of motion similarity. Last, we use the
method of multiple instance retrieval to complete motion retrieval.
Experiment results show that our approaches are effective for motion
data retrieval in agriculture informatisation.
Keywords 3D temporal-spatial; data driven; decision
tree; ensemble; multiple instance; motion retrieval
A07065; Online publication date 4 December 2007; Received and
accepted 10 August 2007
New Zealand Journal of Agricultural Research, 2007, Vol. 50:
583–591
0028–8233/07/5005–0583 © The Royal Society of New Zealand 2007
PDF file of entire paper: Print-quality
(1172K) | screen-quality (517K)
This year's abstracts |
Journal home page |
All abstracts |
Publishing home page
