Paper Info
Title
Grassmannian Sparse Representations and Motion Depth Surfaces for 3D Action Recognition
Abstract
Manifold learning has been effectively used in computer vision applications for dimensionality reduction that improves classification performance and reduces computational load. Grassmann manifolds are well suited for computer vision problems because they promote smooth surfaces where points are represented as subspaces. In this paper we propose Grassmannian Sparse Representations (GSR), a novel subspace learning algorithm that combines the benefits of Grassmann manifolds with sparse representations using least squares loss L1-norm minimization for optimal classification. We further introduce a new descriptor that we term Motion Depth Surface (MDS) and compare its classification performance against the traditional Motion History Image (MHI) descriptor. We demonstrate the effectiveness of GSR on computationally intensive 3D action sequences from the Microsoft Research 3D-Action and 3D-Gesture datasets.
Year
DOI
Venue
2013
10.1109/CVPRW.2013.79
Computer Vision and Pattern Recognition Workshops
Keywords
Field
DocType
manifold learning,grassmannian sparse representations,new descriptor,action recognition,grassmann manifold,traditional motion history image,motion depth surface,optimal classification,computer vision problem,computer vision application,motion depth surfaces,classification performance,minimisation,image classification,classification algorithms,manifolds,accuracy,kernel,principal component analysis,computer vision,dimensionality reduction
Computer vision,Dimensionality reduction,Subspace topology,Pattern recognition,Computer science,Linear subspace,Minimisation (psychology),Artificial intelligence,Grassmannian,Nonlinear dimensionality reduction,Contextual image classification,Manifold
Conference
Volume
Issue
ISSN
2013
1
2160-7508
Citations 
PageRank 
References 
12
0.50
20
Authors
2
Name
Order
Citations
PageRank
Sherif Azary1233.45
Andreas Savakis237741.10