The first question we always get is "How did you do that?" If you were new to virtual reality images when you came to this site, chances were wondering the same thing.
Well, we're going to a shot at explaining this without getting too techie. There are many websites out there on the web that will get into the ugly details of making VR images, so if that is what you are looking for, we suggest doing a search on "making virtual reality panoramas" in your favorite search engine.
This page is for everyone else who are just looking for enough information to impress their friends that know nothing about it at all.
We try to pick locations that will look interesting when they are in VR form and that doesn't always mean picking a spot right in the middle of the room.
We usually spend a lot of time standing in different spots around the area and spinning. We may look a little goofy, but it is the best way to picture how the final VR image will look.
We try to look for a spot that has lines heading into the distance. This could be the edge of a road or a piece of woodwork on the wall. These types of details provide the viewer with a better idea of the size of the area in the finished VR image.
We take the photos one by one, turning the camera on the tripod a little each time. There will be some duplication in the images, but that is a good thing it gives us more choice later on.
After the horizon shots, we turn the camera upwards for a shot of the sky and then down for one of the ground.
One of the biggest problems we face is called "parallax." You can do an experiment to see this problem for yourself. Hold your thumb out in front of you and close one eye. Now open that eye and close the other one. When you switch back and forth the background seems to move.
Below are two combined images. The backgrounds match up fine, but the table in the foreground seems to split on its left edge due to parallax.
Before we put the images all together, we like to do a few touchups to fix problems that may have happened during the shoot.
If there we windows in the room, we take two shots, one exposed for the inside and one for the outside. We put the two photos together so that they look like one:
A. inside B. windows C. combined
Remember earlier when I mentioned that there would duplication in the photos? Well, this is where that comes into play.
Using a special program, we put two photos beside each other and pick the points in them that are the same.
We continue this process all the way around the circle. After we are finished, the program is able to bend the photos to compensate for the distortion of our lens.
While the stitching program can do a lot to make the image look good, there are always a few things that we will have to do by hand.
If there was anything close to the camera, we will have to do some bending to fix the parallax problems.
And sometimes we will remove objects entirely. While I understand the need for blue port-a-potties real reality, we don't need them in virtual reality...
At last, we resize the image to make it download faster on the internet. We add any final gizmos like automatic spinning, popup titles, and hot spots so that the website visitors can click on doors to take a virtual tour.
And a process that took you a few minutes to read probably took us a few hours to do. Making high quality virtual reality images takes quite a bit of patience, time and attention to detail, but, in the end, the reward of wide eyed amazement from viewers is worth it.