Pose Description Based on Natural Relation Sets

Motion capture and pose estimation of human beings is a highly active research area and is related with various applications in many different fields such as 3D character animation, surveillance or human-machine interfaces. A specific problem of these two last applications is the pose and motion estimation of the subject, i.e. understanding which action, among a predefined set, the subject is performing. In this paper, we propose a new high level pose description method based on set of relative body part information, easily understandable by humans, such "above/below" or "between/on each side of". From this pose description, we introduce two kinds of usage: the recognition of a pose described by the user and the detection of poses similar to a set of samples.

[1]  Stefan Carlsson,et al.  Recognizing and Tracking Human Action , 2002, ECCV.

[2]  Saïda Bouakaz,et al.  Real-Time Marker-free Motion Capture from multiple cameras , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[3]  B. Triggs,et al.  3D human pose from silhouettes by relevance vector regression , 2004, CVPR 2004.

[4]  Rémi Ronfard,et al.  Action Recognition from Arbitrary Views using 3D Exemplars , 2007, 2007 IEEE 11th International Conference on Computer Vision.

[5]  Jake K. Aggarwal,et al.  Spatio-temporal relationship match: Video structure comparison for recognition of complex human activities , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[6]  Adrian Hilton,et al.  A survey of advances in vision-based human motion capture and analysis , 2006, Comput. Vis. Image Underst..

[7]  Andrea Kleinsmith,et al.  Recognizing Affective Dimensions from Body Posture , 2007, ACII.

[8]  Trevor Darrell,et al.  Fast pose estimation with parameter-sensitive hashing , 2003, Proceedings Ninth IEEE International Conference on Computer Vision.

[9]  Honghai Liu,et al.  Advances in View-Invariant Human Motion Analysis: A Review , 2010, IEEE Transactions on Systems, Man, and Cybernetics, Part C (Applications and Reviews).

[10]  Marvin J. Greenberg.  Euclidean and non-Euclidean geometries : development and history , 1977 .

[11]  Meinard Müller,et al.  Efficient content-based retrieval of motion capture data , 2005, SIGGRAPH '05.

[12]  Ramakant Nevatia,et al.  Single View Human Action Recognition using Key Pose Matching and Viterbi Path Searching , 2007, 2007 IEEE Conference on Computer Vision and Pattern Recognition.

[13]  Bruno Raffin,et al.  3D Skeleton-Based Body Pose Recovery , 2006, Third International Symposium on 3D Data Processing, Visualization, and Transmission (3DPVT'06).

[14]  Toby Howard,et al.  Real-time markerless human body tracking with multi-view 3-d voxel reconstruction. , 2004 .