Arduino USB Camera

The Arduino® USB Camera is designed for developers, makers, and educators who want to add computer-vision capabilities to their projects. It can be used for face detection, object tracking, gesture recognition, and visual-input applications using local vision-language models.

The camera connects directly to UNO Q and VENTUNO Q via USB and is designed to operate without additional driver installation in supported environments. It provides imaging capabilities for AI inference and computer-vision workflows, and can be used with Arduino App Lab models including YOLO-X, MediaPipe, and Qwen VLM.

A built-in HD microphone is structured to make this USB Camera a complete audio-visual input device in a single unit, ideal for voice-and-vision applications like interactive assistants, touchless interfaces, and multimodal AI projects. Adaptive LED ring lighting with touch-dimming and adjustable color temperature is designed to support image capture across a range of lighting conditions. Flexible mounting options – monitor clip, flat desk stand, or tripod thread – help you position it exactly where your project needs it.

Key benefits

• Plug-and-play AI vision: Designed to work with UNO Q and VENTUNO Q and support computer-vision workflows with minimal setup.
• Adaptive LED lighting for consistent results: Designed to support image capture across a range of environments and lighting conditions.
• Complete audio-visual input in one device: The integrated HD microphone adds voice capture alongside high-fidelity imaging, supporting multimodal AI applications without additional hardware.
• Flexible mounting: Clip it to a monitor, stand it on a desk, or thread it onto a tripod. The USB Camera is built to adapt to your workspace and your project’s field-of-view requirements.
• Purpose-built for the Arduino AI ecosystem: Designed specifically to work with UNO Q and VENTUNO Q and for use with the Arduino App Lab computer vision and AI model library.

Application examples to explore

• Real-time object detection and classification: Use YOLO-X object detection or MobileNet image classification to support face, object, and scene-recognition workflows, while running inference locally on supported devices.
• Gesture-controlled interfaces: Use the camera as the input for MediaPipe gesture recognition to build touchless control systems – translate hand gestures into commands, navigate interfaces without contact, or add an interactive layer to robotics and physical computing projects.
• Vision-language AI and scene understanding: Feed live camera data to a local VLM running on VENTUNO Q. Use local VLMs to support scene description and visual-question-answering applications through on-device processing.
• Multimodal assistants and interactive robots: Combine visual and audio inputs with local AI models to support interactive assistants, educational robots, kiosks, and related multimodal applications.

 
Arduino, UNO, VENTUNO, and the Arduino logo are trademarks or registered trademarks of Arduino S.r.l.