同步操作将从 zzb32/Hand-Keypoint-Detector 强制同步,此操作会覆盖自 Fork 仓库以来所做的任何修改,且无法恢复!!!
确定后同步将在后台操作,完成时将刷新页面,请耐心等待。
This application detects hands in a live video stream.
It was created by integrating two open-source tools:
This application was created to overcome the limitations presented by both the models: Handtrack.js only detects bounding boxes around hands and does not predict finger joints and palm keypoints, whereas HandPose is only able to detect keypoints in a single hand at a time. This application combines Handtrack.js and HandPose to detect the keypoints of multiple hands in a single video frame. The model first applies Handtrack.js to predict the bounding boxes around each detected hand. These bounding boxes are then individually fed to the HandPose model, which detects whether a hand is actually present within the bounding box and, if so, predicts the finger joints and palm keypoints.
The HandTrack.js model often produces bounding boxes that are too tight for the HandPose model to be able to detect a hand within them. Therefore, by default, I scale the bounding boxes by a factor of 2. I have also provided the option to scale the width and height of the bounding boxes according to the user’s preference. Other variables can also be changed, including:
To run locally, navigate to the root directory and run:
python app.py
Then, open a browser and type in the URL http://127.0.0.1:5000/
.
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。