Unverified50% confidenceFactExact time
Google开源的MediaPipe框架的Hand Landmark模型基于MobileNetV2骨干网络，在标准硬件上可达每秒30帧以上的检测速度
1
Sources
50%
Confidence
Long-term
Relevance
7/22/2026
First Seen
Sources
Related Claims
UnverifiedGoogle's MediaPipe provides real-time tracking of 21 hand keypoints and 33 full-body keypoints, suitable for mobile deployment.76% similarUnverifiedGoogle开源的MediaPipe框架提供轻量级手部关键点检测能力，可实时追踪21个手部骨骼节点，通过JavaScript直接在浏览器中运行，无需服务端支持73% similarVerifiedMediaPipe is a cross-platform machine learning framework developed by Google; its face detection module is based on the BlazeFace lightweight neural network architecture and can output 468 facial keypoint coordinates.66% similarUnverifiedGoogle's open-source model Gemma 4 went live on the Cerebras inference platform, running at a speed of 1500+ tokens per second based on 31B open-source weights, achieving approximately 15x acceleration.64% similarUnverifiedGoogle's PoseNet is a lightweight model designed specifically for the browser side, based on the MobileNet architecture.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/594046API
curl https://kongchang.com/api/v1/knowledge/claims/594046MCP
get_claim(id=594046)