360 degrees image sensor
The foundation of the 360Cam is a module that its creators call a “360 degree image sensor”. This module is made of three cameras that can each offer a 180 degree view which are connected to a custom chip (a programmable FPGA in the prototype) that will merge the three video streams into one.
The interesting part about this setup is that the output of the module is not unlike the output of a regular camera module (except for the distortions introduced by the warped field of view), so this could be compatible with any standard image processing workflow. It means that if the company wanted to sell its IP, partners would easily integrate this tech into their own products.
The camera doesn’t need to be connected to a computer, or a smart device. It only has 2 buttons and functions on its own and uses the little LED screen to provide visual feedback to the user. It’s pretty simple: upon turning it on, you will see that it puts itself in photo mode. Pressing a button makes it snap photos every 5 seconds until you stop. The video mode works exactly the same way. When you are done, connect it to a computer and it will show up as a USB drive, just like any other cameras. A photo sample would look like this when viewed in 2D:
Using standard VR viewers
The 360 degree images or videos are compatible with existing VR standards, so it’s possible to view photos and movies in virtual reality (VR) mode and move your phone (or head if you are wearing VR glasses) to see all around you. It is even possible to open files and edit them with regular photo editors like Photoshop or Gimp (free). When viewed in 2D, the 360 degrees images appear somehow distorted or “warped” but you can always see what’s going on. Here’s a 360-photo of me and Richard Ollier, one of the founders of Giroptic, the company behind 360Cam.
Beyond 360 degrees visits
As I said earlier, the first thing that pop to mind is how Realtors could use it to snap home photos in seconds instead of spending time and money to have them stitched and processed with difficult to use software. But you’re probably not a Realtor, so why should you care?
There is a ton of other applications that are made possible with this design. Home surveillance is an obvious one. I’m not sure if you have noticed, but most $70 Internet cameras have ridiculously narrow fields of views. The coolest one that I have played with (in the sense that it’s super-easy to use) is the Dropcam Pro, but it costs $199. Samsung has some interesting alternatives, but I have yet to open that box.
For an estimated $250, the 360Cam could deliver a 360-degree view and Giroptic told me that they could give it recording and remote controlling capabilities close to what Dropcam has, thanks to a 3rd party. What about wires and mounting? Just look at the light bulb mounting: very smart.
Semi-pro/Pro security is another one. Because you can cover much more area for the price, it is possible to cover much more ground in a cost-effective way. That said, this won’t replace the super high-end setups with zooms and. While visiting the Sony HQ in Tokyo, they showed me one of their surveillance system that could read a license plate more than a mile away – but again, the price is in a different universe.
Built-in Modularity for future hacking
The 360Cam has been built with native modularity in mind. The bottom of the camera can be redesigned to fit a particular need. When I saw the bulb mount, I thought that it would be possible to create a rugged version with a round and heavy bottom to throw it and have it land vertically every time. Maybe a SWAT team could use some 360-degree live video stream…
As it is, the 360Cam design is also waterproof. Its inventors have even added some little lens caps, not to protect it from water, but to make sure that the images won’t be distorted by the refractive properties of water.
"360CAM HAS A LOT OF POTENTIAL"The 360Cam has a lot of potential, and of course, this is a prototype at this point, so while I have been able to see that it works, I haven’t had a chance to see the full user experience which includes using it in a real-world situation and on a daily basis, which is where any user-interface friction would happen.
That said, I don’t see a reason why its team could not pull it off. So far, the fundamentals work well enough, and the fact that the camera works with a programmable FPGA chip shows that there is a lot of potential to make it much faster with a dedicated chip (FPGAs are slower than production chips, but can be re-programmed, so it’s great for developers).
The Kickstarter campaign starts tomorrow, so I’ll update this post with a link for those of you who want to know more about it.