Pose Estimation and Vision

Data gathered from various sensor modalities  can be used to model and classify complex 3D scenes.  

 

Hand Gesture Classification

We have used 3D sensor data  for pattern classification of gestures. mobile int

 

  • Di Benedetto, Anna; Palmieri, Francesco A. N.; Cavallo, Alberto; Falco, Pietro, "A Hidden Markov Model-Based Approach to Grasping Hand Gestures Classification," in S. Bassis, A. Esposito, F.C. Morabito, and E. Pasero, Advances in Neural Networks, (WIRN 2015), vol. 54, pp. 415-423, ISBN: 978-3-319-33747-0; DOI: 10.1007/978-3-319-33747-0_41; Springer Science 2016.

 

Pose Estimation from Point Clouds

 Point clouds gatherd from the Kinect sensor are used for pose estimation.  mobile int

 

  • P. Coscia, F. A. N. Palmieri, F.Castaldo, A. Cavallo, "3-D Hand Pose Estimation from Kinect's point cloud using Appearance Matching," in S. Bassis, A. Esposito, F.C. Morabito, and E. Pasero, Advances in Neural Networks, (WIRN 2015), vol. 54, pp. 37-45, ISBN: 978-331933746-3; DOI: 10.1007/978-3-319-33747-0_4; Springer Science 2016.

 

Planar Surface Estimation with Kinect

 Point clouds gatherd from the Kinect sensor are used for modeling planar surfaces. mobile int

 

  • F. Castaldo, V. Lippiello, F. A. N. Palmieri and B. Siciliano, "Real-time Estimation of Planar Surfaces in Arbitrary Environments using Microsoft Kinect Sensor," Proceedings of 17th International Conference on Image Analysis and Processing ICIAP 2013, Sept. 9-13, Naples, Italy, pp. 552-561, Part I, Springer 2013 Lecture Notes in Computer Science, ISBN 978-3-642-41180-9. , 2013