The Shashin Go series of platforms is based around the idea of a wearable camera. It is intended to augment existing devices (phone, computer, etc.).
Guiding principles:
The Shashin Go Concept 1 is an Open Source Hardware design for AI data capture. · Open Source software provided gives the user complete control over the camera resolution and data format · Software is available on Github and can be reflashed with alternative firmware using either Arduino or the Sony Spresense SDK · Ships in camera mode, able to capture photo streams and save data to an SD card · Alternative firmware provides Tensor Flow Lite based Person Detection
Sony Spresense platform This platform features a 6-core MCU and a 5MP camera. It is well suited for IoT and AI applications, yet has the simplicity of the Arduino ecosystem.
Concept 2 is known as RearViewsion, features a wearable Wifi Camera with a companion application for the VisionPro. It gives the visionpro the missing external camera, in this case, geared towards rear facing vision. It was a winner of the "Best Hardware" category at VisionDevCamp: https://visiondevcamp.org/visiondevcamp-winners/
The plan for this hardware is to get it for sale on tindie in several variations as I try out different design concepts. There are hundreds of possible features.
I would like to test the market to see what is the most useful and cost offective to build. I expect different offerings in the $30-$100 range, based on enclosure,
memory, SD storage, applications, etc.
Source Code: https://github.com/IoTone/ShashinGo/tree/main/Concept2/RearViewsion