This is my video response to “Lenny Lipton discusses the REAL D digital 3D system”.
I beg for more videos about stereoscopy, and I have a question about triple-shuttering.
This video is in 3D.
Archive for February, 2007
View the Video
Also uploaded to YouTube.
Hopefully, I will get better and better at this, but this video shows the two most important things, IMO, for making good 3D anaglyph videos for YouTube, LiveVideo, or other Flash video sites.
The first thing to do is minimize the depth. It only takes a little bit to get the job done. An “optimum” amount is actually a lot less than most people probably realize, anyway, and “a little too much” will absolutely kill your videos when you try to compress them.
The other most important thing to do is properly set your screen parallax to a point that works the best. If done “incorrectly”, the compression can make the video unwatchable.
Keep in mind that you have to compress your video to begin with for video sites like YouTube, so that the file size is small enough for you to be able to upload, and then the video site will again compress that video to Flash. This double-whammie of compression will guarantee that a “poorly designed” anaglyph will be totally destroyed in the process. Only the most carefully designed anaglyphs will survive.
This video is in 3D.
Update! This project file and explanation produces a tweakable “Half Color” anaglyph, i.e. the left channel is eventually converted to grayscale (or actually desaturated). Nov 4, 2007, I came up with a much easier way to optimize anaglyphs with After Effects by using a Channels Mixer filter. I suggest you try it! Please read this post:
http://www.puppetkites.net/blog/archives/95
This Adobe After Effects (ver. 6.5 Pro) project file is how I convert a separate stereoscopic left and right video to a parallel version and/or a color anaglyph version, using my own special anaglyph conversion process, which can be tweaked differently for every 3D video segment. You can render either the Parallel or the Anaglyph comp, e.g. you can use the anaglyph version just for previewing the stereoscopic alignment, if you wish, and render just the parallel version for free-viewing or real-time conversion to other stereoscopic formats with suitable stereoscopic software.
Here is the project file:
Left and Right to Parallel and Anaglyph
Replace the “missing video files” with your own videos, and if you are importing something other than 720×480, 60i NTSC videos, you’ll have to change the video settings and the size of the over-sized black overlays in the “masked” comps to fit your own situation. I also convert the resolution to 640×480 square pixels in this project.
Notice the videos are imported at 29.97fps and the fields are separated (Interpret Footage), but the first Left and Right comps are set to 59.94fps (twice the frame rate). This allows you to shift either the left or right video in one of those comps 1/60th of a second on the timeline for the best possible time-sync, if necessary. The next “Masked” comps, along with all the other following comps, are set back to the original frame rate, and any time-syncing will be reflected in all those comps.
Here’s my coolest pair of 3D glasses, and the brief story behind them. Let’s see your coolest pair, and hear a little bit about them… I know I probably don’t have the very coolest pair…
This video is in 3D.
Re: The Value of Video Comments
This is my video response to The Value of Video Comments.
This video is in 3D.