Datasets
Standard Dataset
K-CEGR: Enable Cross-Domain High Precision Gesture Recognition with Kinect
- Citation Author(s):
- Submitted by:
- zhixiong Yang
- Last updated:
- Sat, 12/03/2022 - 02:15
- DOI:
- 10.21227/yrfr-sg43
- License:
130 Views- Categories:
- Keywords:
0 ratings - Please login to submit your rating.
Abstract
This is the continuous Chinese and English gesture data of 14 Chinese and 4 English languages, respectively “不”，“程”，“刀”，“工”，“古”，“今”，“力”，“刘”，“木”，“石”，“土”，“外”，“中”，“乙”，“can”，“NO”，“Who”，“yes”.
Instructions:
Through 10 different users (5 men and 5 women), (ji) kinect collected continuous Chinese and English gesture data about 14 Chinese and 4 English languages in three different environments through M-YOLOv5 algorithm, which are respectively “不”，“程”，“刀”，“工”，“古”，“今”，“力”，“刘”，“木”，“石”，“土”，“外”，“中”，“乙”，“can”，“NO”，“Who”，“yes”, After data processing, there are 28000 "yes" gestures, 70% of which are training sets. 30% as test set.
