Go With The Flow: Hand Trajectories in 3D via Clustered Scene Flow
Hadfield, Simon J. and Bowden, Richard (2012) Go With The Flow: Hand Trajectories in 3D via Clustered Scene Flow In: International Conference on Image Analysis and Recognition, 25-27 June 2012, Aveiro, Portugal.
![]()
|
Text (Camera ready copy of paper.)
175.pdf - Accepted version Manuscript Available under License Creative Commons Attribution No Derivatives. Download (2MB) |
|
![]() |
Video (Video of 3D tracking of sign language, in a 2 view cluttered setup.)
supplementary1.avi - Supplemental Material Available under License Creative Commons Attribution No Derivatives. Download (4MB) |
|
![]() |
Video (Video of 3D head tracking of sign language, in a 3 view uncluttered setup.)
supplementary2.avi - Supplemental Material Download (5MB) |
Abstract
Tracking hands and estimating their trajectories is useful in a number of tasks, including sign language recognition and human computer interaction. Hands are extremely difficult objects to track, their deformability, frequent self occlusions and motion blur cause appearance variations too great for most standard object trackers to deal with robustly. In this paper, the 3D motion field of a scene (known as the Scene Flow, in contrast to Optical Flow, which is it's projection onto the image plane) is estimated using a recently proposed algorithm, inspired by particle filtering. Unlike previous techniques, this scene flow algorithm does not introduce blurring across discontinuities, making it far more suitable for object segmentation and tracking. Additionally the algorithm operates several orders of magnitude faster than previous scene flow estimation systems, enabling the use of Scene Flow in real-time, and near real-time applications. A novel approach to trajectory estimation is then introduced, based on clustering the estimated scene flow field in both space and velocity dimensions. This allows estimation of object motions in the true 3D scene, rather than the traditional approach of estimating 2D image plane motions. By working in the scene space rather than the image plane, the constant velocity assumption, commonly used in the prediction stage of trackers, is far more valid, and the resulting motion estimate is richer, providing information on out of plane motions. To evaluate the performance of the system, 3D trajectories are estimated on a multi-view sign-language dataset, and compared to a traditional high accuracy 2D system, with excellent results.
Item Type: | Conference or Workshop Item (Conference Paper) | |||||||||
---|---|---|---|---|---|---|---|---|---|---|
Divisions : | Faculty of Engineering and Physical Sciences > Electronic Engineering > Centre for Vision Speech and Signal Processing | |||||||||
Authors : |
|
|||||||||
Date : | 25 June 2012 | |||||||||
DOI : | 10.1007/978-3-642-31295-3 | |||||||||
Uncontrolled Keywords : | hand tracking, 3d, scene flow, particle filter, scene particles, sign language, multi view | |||||||||
Additional Information : | This is an Open Access article distributed under the terms of the Creative Commons Attribution License (No derivatives)(http://creativecommons.org/licenses/by/2.0). The original publication is available at www.springerlink.com | |||||||||
Depositing User : | Simon Hadfield | |||||||||
Date Deposited : | 07 Sep 2012 14:56 | |||||||||
Last Modified : | 19 Jun 2018 06:03 | |||||||||
URI: | http://epubs.surrey.ac.uk/id/eprint/380513 |
Actions (login required)
![]() |
View Item |
Downloads
Downloads per month over past year