Boxer3D Ports Meta's 3D Object-Lifting Pipeline to LiDAR-Equipped iPhones
Sign in✓ Signed in
An independent developer has published Boxer3D, a Swift prototype that combines YOLO11n, iPhone LiDAR and ARKit to turn 2D detections into gravity-aligned, 7-DoF oriented 3D bounding boxes rendered with SceneKit. The project moves BoxerNet inference onto supported iPhones, although it is a narrower, single-frame adaptation rather than a complete mobile port of Meta Reality Labs Research's open-vocabulary, multi-view Boxer system.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article…