Combining on-device AI, high-speed depth sensing, and a range exceeding 20 metres, the new camera aims to simplify next-generation robotic vision.
A new AI-enabled 3D depth camera has been introduced for robotics and industrial automation, combining on-device AI processing, long-range depth sensing exceeding 20 metres, and high-speed image capture in a compact system. By performing AI inference directly on the camera, it reduces latency and system complexity while simplifying computer vision development for robots, drones, warehouse automation, industrial inspection, autonomous navigation, and collaborative robotics.
Some of the key features include:
- On-device AI processing
- Depth sensing beyond 20 metres
- Object detection from as close as 10cm
- Up to 90fps depth capture
- 1280 × 960 depth resolution
- 120° × 100° field of view
- Global shutter sensors
- Dual infrared projectors with integrated IR filters
- Synchronized RGB and depth streaming
- Around 75% lower bandwidth through depth compression
The camera integrates a depth engine, image signal processor, AI acceleration hardware, and a quad-core Arm processor, enabling image processing and AI inference directly on the device instead of relying on an external processor.
It also supports synchronized RGB and depth streams, on-camera stream alignment, and depth compression technology to reduce data bandwidth during high-frame-rate transmission. Future software updates are expected to add capabilities such as person detection, visual-inertial odometry, occupancy-grid generation, auto-calibration, face detection, and enhanced depth-processing features.
