Title:
GESTURE RECOGNITION METHOD, VR ANGLE OF VIEW CONTROL METHOD AND VR SYSTEM
Document Type and Number:
WIPO Patent Application WO/2020/073245
Kind Code:
A1
Abstract:
A gesture recognition method, a VR angle of view control method and a VR system. The gesture recognition method comprises: acquiring depth information (310); acquiring spatial point cloud information according to the depth information (320); determining a target zone in the spatial point cloud information, the target zone referring to a zone containing hand point cloud information (330); generating a plane image corresponding to the target zone (340); extracting edge points of the hand in the plane image (350); determining the number of fingers of the hand according to the edge points of the hand (360); determining a gesture of the hand according to the number of fingers (370). By means of the described method, a gesture is detected and recognized by means of geometric morphological analysis without needing to rely on sample learning algorithms such as machine learning, which may guarantee a higher gesture recognition rate while effectively reducing the operation volume of gesture recognition, and satisfy the application requirements of a low power consumption, low delay and low hashrate platform.
Inventors:
ZHENG XIN (CN)
Application Number:
PCT/CN2018/109698
Publication Date:
April 16, 2020
Filing Date:
October 10, 2018
Export Citation:
Assignee:
AUTEL ROBOTICS CO LTD (CN)
International Classes:
G06T7/20; G06F3/01
Foreign References:
| CN102982557A | 2013-03-20 | |||
| CN102982557A | 2013-03-20 | |||
| CN104914993A | 2015-09-16 | |||
| CN104317391A | 2015-01-28 | |||
| CN108230383A | 2018-06-29 |
Attorney, Agent or Firm:
LIUJIA CHINA IP LAW OFFICE (CN)
Download PDF: