Code repository for the Open Source 360 Camera Project

Brought to you by Photon Labs: photonlabs.tech

See us on hackaday.io

This repository contains code to capture images and eventually process the raw images for panoramic or spherical panoramic stiching.

• This was created using 2 images captured with a 180 degree rotation (I only have 1 of the 220 degree lenses currently)
• Created with the free version of PTGui
• Can be viewed using PTview, google street view app or other 360 spherical viewers *Taken with the LS-40180 fisheye (lower quality, but has an IR filter so color is correct)

View on Google Stree View

Dowload full image here: http://imgur.com/TNTswUM

• Taken with the 1.17mm 220 degree lens DS13F117M12 from Day Optics
• Camera white balance is off due to no IR filte on lens

Download full size image here: http://imgur.com/FxKGasV

The entrance of consumer grade and professional 360 spherical cameras to the market will change the way people capture and interact with digital images and video. While cameras like the Ricoh Theta, Samsung Gear 360 and Facebook Surround 360 cameras will meet consumer and professional needs, no platform is available to the maker, hacker, or educator.

This goal of this project is to develop a hardware kit that is open, hackable, adaptable, and extensible to open this incredible technology to everyone.

As detailed on the hardware page, Our key principals are as follows:

• Open source hardware and software to encourage educational use and a developer community
• The primary hardware is to be off the shelf (lenses and image sensor PCBs)
• Simple mechanical mounting, can be assembled without any specialized tools or equipment
• Compatible with mobile computing platforms such as Raspberry Pi or similar
• Cost to be similar to a consumer level 360 camera such as Ricoh Theta ($250-$350)
• Performance to be adequate to create 360 panoramas

To enable portable imaging, we have developed and tested the cameras on a Raspberry Pi 2 B utilizing OpenCV libraries to control and capture images from the cameras. The USB 2.0 cameras from ELP that we selected us the Sonix SN9C292A video controller with USB Video Class compatibility.

To see examples of images captured with the hardware see:

• Field Testing

Code description:

File | Description --- | --- | --- Live2CameraDisplay.py | Live display from two cameras at 800 x 600 resolution. Quits when 'q' is pressed FullSizeSnapShot.py | Live display and captures full res images from both cameras when 'q' is pressed FucusUtility.py | Utility for setting lens focus. Crops central region and zooms display. Hit q to close

Current dependancie:

• http://opencv.org

Depedancies in progress:

• https://github.com/ktossell/libuvc.git
• https://github.com/pupil-labs/pyuvc.git (Python UVC)