MoveNet 3D – Realtime 3D Pose Tracking – Published By NatML

Perform realtime 3D pose tracking with the MoveNet machine learning model. This package requires the open source

Keywords:

ARFoundation, ARKit, avatar, machine learning, motion capture, natml, arcore, tflite, Computer Vision, mocap, Augmented Reality, tensorflow, mediapipe, pytorch, body tracking

Short Description:

Perform realtime 3D pose tracking with the MoveNet machine learning model. This package requires the open source

Rating:

Discount: None

Asset Title: MoveNet 3D – Realtime 3D Pose Tracking

Publisher: NatML

Category: tools, integration

More Details about this asset:

Perform realtime 3D pose tracking with the MoveNet machine learning model. This package requires the open source NatML, NatMLX, and ARFoundation libraries to run the machine learning model. Features include:


+ Bare Metal Performance. The MoveNet 3D predictor is implemented with NatML, which takes advantage of hardware machine learning accelerators, like CoreML on iOS and macOS, NNAPI on Android, and DirectML on Windows.


+ Extremely Easy to Use. The MoveNet 3D predictor accepts an AR camera and depth image and returns a 3D body pose with 17 keypoints along with the confidence score. See more on NatML Hub.


+ Cross Platform. The MoveNet 3D predictor supports Android and iOS alike, allowing you to develop once and deploy on both platforms. You no longer have to rely on platform-specific API’s like ARKit’s body tracking.


+ Augmented Reality. This predictor uses depth data from ARFoundation, so it has first-class support for augmented reality apps.


+ Motion Capture and Avatars. The 3D pose from MoveNet 3D can be used to create motion capture demos and even drive virtual avatars!


+ Lightweight Package. This package contains the predictor scripts, whereas the ML model will be downloaded at runtime from NatML Hub and cached onto the device, reducing the app size significantly.


***NOTE***

This predictor can only be used on devices that support AR depth (human or environment depth), see technical details for more info.

***********


Find other ML models on NatML Hub.


Join the NatML community on Discord.


Check out the NatML documentation.


See more NatML projects on GitHub.


Read the NatML blog.


Discuss NatML on Unity Forums.


Contact us at hi@natml.ai.

Leave a Reply