LiteRT.js Mocap enables browser 3D motion capture

// 1h agoOPENSOURCE RELEASE

LiteRT.js Mocap enables browser 3D motion capture

LiteRT.js Mocap is an open-source demo repository showcasing Google's LiteRT.js runtime for high-performance, on-device machine learning in the browser. The application performs real-time 3D pose estimation from a webcam feed to drive an avatar using Three.js, supporting WebGPU acceleration and Blender export.

// ANALYSIS

Google's rebranding and optimization of TensorFlow Lite into LiteRT.js signifies a major step toward making high-performance machine learning standard in web applications.

* Near-Native Browser Performance: Utilizing WebGPU, WebAssembly, and WebNN helps close the performance gap between web apps and native applications.

* Decoupled Frameworks: By supporting JAX, PyTorch, and Keras models under the LiteRT umbrella, developers are no longer locked into the TensorFlow ecosystem.

* Interactive Application Demo: Standardizing features like Blender export makes this repository a practical utility for content creators, not just a technical showcase.

// TAGS

litert.js-mocapllmmotion-capturewebgputhree.jscomputer-vision3d-avataropen-source

DISCOVERED

1h ago

2026-07-17

PUBLISHED

1h ago

2026-07-17

RELEVANCE

7/ 10

AUTHOR

Better Stack

// KEEP READING

More AI developer news from the feed

EXPLORE FULL FEED

UPDATE36m ago

Orca Mobile launches agent chat UI

Orca has released its Chat UI for Orca Mobile in beta for iOS and Android, allowing developers to monitor and control desktop AI coding agents remotely. Developed with RunFusion, the update introduces a free mobile relay service that eliminates the need for a Tailscale setup.

FUNDING42m ago

After Labs emerges from stealth with funding

After Labs is a newly unveiled AI research lab focusing on the development of efficient fluid intelligence. The startup has officially come out of stealth after securing funding, drawing congratulations from prominent AI figures including François Chollet for founders Clem and Matt.

MODEL1h ago

Thinking Machines' Inkling model hits OpenRouter

Thinking Machines Lab has made their new open-weights Mixture-of-Experts (MoE) model, Inkling, available on OpenRouter. The model features 975 billion total and 41 billion active parameters, supports a 1 million token context window, and provides controllable reasoning across text, images, and audio.