LivePose offers a simple way to run computer vision models on live video and send the results to any OSC-compatible application. With no programming required, load a model, select your camera, and start streaming data. This release uses ONNX, an open standard for machine learning models, providing a lightweight, cross-platform GUI application.
- 🧠 Load
.onnxmodels (BlazePose, YOLOv8 Pose, ResNet) - 📷 Automatic camera detection (plug and play)
- 🎬 Live video preview
- 📡 Configurable OSC output
Connect your output data to Pure Data, Max/MSP, Processing, p5.js, TouchDesigner, SuperCollider, openFrameworks, or any software that receives OSC. Useful for interactive installations, live performance, prototyping, and experimentation. Check awesome-creative-coding for more tools and resources.
Developed by the Société des Arts Technologiques, built on ossia score.