3 results about "Right palm" patented technology

The utility model discloses a novel bunt tide, relates to musical instrument technical field, including the tube body, four tuning holes are seted up in the tube body front side bottom, the replaceable contact mouth is provided with three clamping blocks in the array of tube body top end outer wall bottom end outer wall, and the fixed connection of limit stop is provided with in the tube body outer wall close to the position of clamping block, the protection ring is provided with in the tube body bottom end outer wall, the auxiliary mechanism is provided with in the tube body outer wall close to the position of tuning hole, and the auxiliary mechanism includes left -hand auxiliary block and right -hand auxiliary block. The utility model discloses design structure is reasonable, and the left hand index finger and middle finger are placed respectively on the both sides of limit ring B, and the left hand index finger is positioned and is guided to the groove A of placing, and right hand index finger and middle finger hold the tube body and are placed below limit ring C, and right hand thumb is placed in the groove B of placing on limit ring C under the guidance of limit ring C, and contact block A and contact block B support left and right palm while holding the tube body, have the advantage that the hand of player is attached.

The application discloses a palm vein ROI extraction method and system based on a light network, and the method comprises the following steps: first, inputting a palm vein image into an improved HRnet deep learning key point detection model to position key points; then, positioning a palm vein region of interest and a selected region position through key point coordinates obtained through geometric operation and fine adjustment; in order to correctly select the palm vein region of interest, the left and right palms are distinguished; finally, an affine transformation algorithm is used to obtain a final region containing the vein. Through the method, redundant network branches are deleted, the number of convolution channels is reduced, and a deep separable convolution mode is adopted for a residual block structure to reduce the network parameter quantity; while the key point detection accuracy is considered, the model calculation amount is greatly reduced, and the palm vein ROI extraction speed is accelerated. The application can be widely applied to the field of biometric recognition technology.

The invention relates to the technical field of virtual reality gesture recognition, and provides a Meta XR two-hand cross gesture recognition method based on a virtual point ray, and the method comprises the steps: operating a target application in a virtual reality environment of Meta XR equipment; calling a gesture recognition interface provided by the virtual reality environment, collecting hand skeleton information data of a user, and extracting skeleton nodes of left and right palm centers; taking palm center skeleton nodes of the left hand and the right hand as virtual points, taking forward directions of local coordinate systems of the virtual points as directions of palms, adopting a vector point multiplication method to perform vector operation on the forward directions of the left hand and the right hand, and when a point multiplication result is smaller than a preset threshold value, judging that a double-hand cross gesture exists, and triggering preset interaction feedback in the virtual reality environment. According to the method, the virtual point ray is generated based on the direction information of the palm center skeleton node without depending on the preset gesture configuration of Meta XR, and the two-hand cross gesture is judged through the vector point multiplication operation, so that the accuracy of two-hand interaction recognition is effectively improved.