The invention discloses a hand posture recognition method, a storage medium and a vehicle. The method comprises the steps that a time sequence hand 3D image is acquired; performing feature extraction on each hand 3D image in the time sequence hand 3D images to obtain time sequence hand 3D skeleton key points; and inputting the time sequence hand three-dimensional skeleton key points into a pre-trained hand posture recognition model, and recognizing to obtain a hand posture. According to the method, hand posture recognition is carried out based on multiple frames of images, more diversified gestures can be recognized, more gesture interaction functions can be achieved, meanwhile, few computing resources are needed, the applicability is high, and the method has the great cost advantage when newly-added gestures exist.
本发明公开了一种手部姿态识别方法、存储介质以及车辆,方法包括:获取时序手部3D图像;对时序手部3D图像中的各手部3D图像进行特征提取,得到时序手部三维骨骼关键点;将时序手部三维骨骼关键点输入至预先训练好的手部姿态识别模型,识别得到手部姿态。该方法基于多帧图像进行手部姿态识别,能够识别出更多样性的手势,有利于更多手势交互功能的实现,同时且需要的计算资源少,适用性强,在有新增手势时有较大的成本优势。
Hand posture recognition method, storage medium and vehicle
手部姿态识别方法、存储介质以及车辆
2023-02-03
Patent
Elektronische Ressource
Chinesisch
Real-time hand posture recognition using range data
British Library Online Contents | 2008
|Exploring Arm Posture and Temporal Variability in Myoelectric Hand Gesture Recognition
BASE | 2018
|Seat posture adjusting method and device, vehicle and storage medium
Europäisches Patentamt | 2023
|Lifting appliance posture recognition method, device and equipment, storage medium and crane
Europäisches Patentamt | 2023
|Driving sitting posture recognition method and device and computer readable storage medium
Europäisches Patentamt | 2023
|