Video file and display resolutions for VR experiences

I was trying to figure out an ideal video resolution to record VR video in, for my HTC Vive VR headset. By ideal, I mean what video file resolution would take full advantage of the display’s resolution.
I am asking myself this question because I don’t think the current state of technology has a good answer and I want to know what recording resolution do I need to max out the amount of recorded detail.

It’s display resolution is 1080×1200 per eye, with a field of view of 110deg side to side and 100deg up and down.
That means that side to side, to cover 360 deg the ideal resolution should be 3534px (360/110*1080).
Up and down, to cover 180deg, the per eye resolution should be 2160 px (180/100*1200)
So the ideal 360deg file resolution should be 3534 x 2160px per eye. For stereo video, the vertical resolution needs to double, so it beomes 3534 x 4320. That’s very close to 6k resolution (4,992 x 3,744, which is 18MegaPixels)

Everyone that’s tried a VR headset in 2017 will attest to the the fact that the display resolution is rather poor. But even with this ‘poor’ display resolution we need 6k (18 MegaPixel) video to take full advantage of it. 6k video files are massive and working with them is not a simple task from a computational perspective. Only the very best/most powerful machines can handle smooth editing at these resolutions. 7680×4320

If we want to increase that display resolution to get a less pixelated image in the headset display, we are also going to have to increase the recorded video resolution. So even a modest 1.4x increase in headset display resolution on both axis (which results in a 2x total resolution increase) means that we need 10k video to use the display’s full potential. Going to 2x increase in display resolution on both axis (so a 4x increase in total resolution) means… I don’t think we even have resolutions that go up that high… 20k resolution?
I could not even speculate on what kind of internet connection and bandwidth you would need to stream that kind of video file from youtube.

All this being said, while a 4 times increase in resolution of the display is significant, it does not come close to full potential of what the eye can see and distinguish.

The current 4k video streams available on youtube do not contain enough information to take full advantage of the display resolution in today’s generation of headsets.

No Comments Video, VR

Connecting phone to Vuze Camera

I had some trouble connecting my Android phone to the new Vuze Camera from HumanEyes. I contacted their support department, and they offered some help, but to follow their instructions to connect I would have to reset the WiFi settings every time.

The problem I was having is that even though I would connect my device to the WiFi network that the Vuze Camera created, the software would not be able to connect to display the live feed and access the settings.

After playing around some more with the camera, I figured out what’s going on.

The software on your device connects through WiFi to the camera. The software likely tries to connect to some preset IP address of the camera. If the device you have doesn’t have a cellular connection, this method of connection works every time with no hassles.

However if you’re connecting from a phone that already has a cellular network connection, things change. When my phone connects to the Vuze network, it informs me that the Vuze network has no internet access, and proceeds to use the cellular network for continued internet access. Because of this, one of two things is likely happening:

1.  The IP address the Vuze software on the phone is trying to access likely gets directed out to the internet… making the Vuze inaccessible to the phone.
2. The subnet that the cellular network DHCP service is supplying the phone makes the Vuse camera IP address not addressable from the phone’s IP address.

So in order to always be able to connect to the Vuze camera from my phone, I have to:

1. Put the phone in Airplane Mode

2. Enable ONLY Wi-Fi on the phone, and connect to the Vuze network.

Vuze software will now reliably connect to the camera every time. I would like to still have cell network connection while connected to the Vuze camera, but I don’t know if this is something that can be fixed in software.

No Comments Video

Vuze Camera RAW resolution

Spherical video is tricky. The process of stitching together multiple images to create a equirectangular projection image doesn’t give you an exact image resolution. So what resolution should I be outputing my video to? Too low a resolution and you’re loosing detail. Too high a rendered resolution and you’re wasting bandwidth as no more details are created when an images is scaled up.

In using the new Vuze Camera from HumanEyes, I wanted to figure out what the ideal rendered output should be to maximize image quality and minimize file size waste. It’s easy to output to 4096 x 4096 px and be done with it, but that resolution of is not easily playable by most devices today, and it may be a waste of bandwidth.

So the goal is to get a ballpark idea of what the final rendered resolution should be based on the data recorded by each sensor, in order to retain as much detail as possible from the raw footage to the youtube file as possible, in order so the viewing experience is as sharp as possible when viewed in 3D 360º, without wasting extra file size.

The resolution I came up with is 3200 x 2880 pixels. Read on to find out how I came to that conclusion.

Read More

No Comments Video, VR

Video Workflow (May2017)

Getting into video work, requires a workflow to be efficient. This is the current iteration of the workflow, which I’m sure will change over time.

I’m still undecided if I should shoot/edit 4k, or if I should just do 1080p material. The hesitation stems from the fact that it takes a lot of computing power to render effects (especially stabilization) on 4k footage.

In camera shoot in S-log most of the time. Especially if I need to bump up the ISO. (talking about the A7Rii camera)

Bring footage into the computer and convert it to Cineform on ingest. This codec is much quicker at scrubbing through and working with then the H264/mp4 stream that comes from the camera.

WHen stabilizing video in AE using masks in the Warp plugin, make sure the Opacity of the masks is set to 0%. If’ it’s left at the default 100%, rendering time is incredibly long!

No Comments Video