Verified70% confidenceFactExact time
MediaPipe is Google's open-source machine learning pipeline framework; its gesture recognition module can track 21 hand keypoints at over 30 frames per second on ordinary smartphone cameras, with latency as low as tens of milliseconds.
4
Sources
70%
Confidence
Long-term
Relevance
7/19/2026
First Seen
Sources
Related Claims
UnverifiedMediaPipe's Hand Tracking module can recognize 21 key landmarks on a hand in real time from camera footage.82% similarUnverifiedLightweight models such as MoveNet and BlazePose in Google MediaPipe can run pose estimation in real time at 30fps or higher on mobile devices67% similarUnverifiedThe project is a pose-based teleoperation system that captures human body pose in real time via an ordinary webcam and maps it onto the joints of a quadruped robot.65% similarUnverifiedA developer built a browser-based gesture interaction application on top of MediaPipe's HandLandmarker that supports pinch-to-zoom, open-palm rotation, and point-and-dwell selection, running entirely on the client side without uploading video.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/558252API
curl https://kongchang.com/api/v1/knowledge/claims/558252MCP
get_claim(id=558252)