DVD Player Displays 2D Movies in 3D

Anonymous Writer writes "A company called Dynamic Digital Depth that wants to bring 3D television and movies to the mainstream claims to have developed a system that allows you to watch current 2D DVDs in 3D. They claim the TriDef DVD Player uses image analysis methods, developed by the company for their 3D content conversion service, to convert 2D video to 3D in real-time based on 3D depth cues in the original movie. It is the same company that produced the TriDef Movie Player software for the Sharp Actius R3D3 autostereo display notebook. "

Re:Dubious

[Anonymous Coward]

there is an easy answer, no one really wants to buy what nature already gave you, but if you can make it sound and feel better then what you already have, then people will buy it. (see marketing, capitalism, snake oil)

It is possible..

[StacyWebb]

to capture the information based on filters. This filter would "pre" read the clip information and then convert the "blurry" or background images and then move the primary "focal point" image further towards the "front" of the clip.

Re:Video Games

[cardshark2001]

Imagine if this could work for video games.

There are a couple of different stereo 3d shutter glasses that work for games on your computer which are already 3d, like quake3 or unreal.

They work by cutting the effective frame rate in half, and rendering each frame twice from a different perspective, and flashing the image into each eye on alternate frames.

Not sure if I explained that very well, but I've seen the "Revelator" (now defunct) in action, and I must say the results, while not perfect, are very impressive.

More detail

[Overt Coward]

Article containing more info [digitmag.co.uk]

Choice quote:

But the result isn't quite like viewing something truly filmed in 3D. Most of the 3D effects are "from the screen backwards, (with) no off-the-screen effects," Harman says. This could be a disappointment to aficionados of 1950s guilty-pleasure flicks, who know that the whole point of watching a 3D movie is to see various objects (mostly monsters) popping out of the screen.

Lots of Research on this

[Roger_Wilco]

Binocular disparity only works out to a few metres distance. Beyond that you use different cues. Consider some papers by my supervisor, for example: A laminar cortical model of monocular and binocular interactions in depth perception [journalofvision.org], Neural Dynamics Of 3-D Surface Perception: Figure-Ground Separation And Lightness Perception [bu.edu]

A friend of mine had this idea

[Orion Blastar]

to take the red and green pixels and move them slightly off a bit like a real 3D movie does. The shadows and other details will be seen as 3D.

I recall that there was photo editing software that did this to 2D picture images, so it is possible to do it to a 2D movie in real-time should the CPU be fast enough to do it.

To quote that Wendy's lady from the 1980's "Where's the beef?" I searched those sites and could not even find a demo! Is it vaporware or real?

Re:Independent review?

[baxissimo]

I don't know it's the same one, but a couple of summers ago a company came to the place where I was working and gave a demo of their "revolutionary technology to turn 2D movies into 3D movies". I went to see it. I was spectacularly unimpressed. They were doing something to the edges of moving objects, but whatever it was it wasn't 3D. If you've ever looked at two slightly different images with your two eyes, then you know that sort of shimmery effect where there are differences in the images? Your brain sort of wants to interpret those differences as 3D somehow even if they aren't really. I think that's what they were taking advantage of, and hoping that if things looked "funky" in that way people would jump up and down and say "ooh ahh it's 3D!". But anyone who has a fair amount of technical savvy would not be impressed by the system I saw.

But that was a few years ago, maybe they've made progress. But I doubt it. I'm guessing the only breakthroughs they've had are with their marketing department. Most serious depth extraction algorithms still barely crawl on multi-GHz machines, and they still don't do a very good job. If they can do this in real time then they definitely aren't extracting any sort of real depth. Just playing mind games.

Re:Dubious

[Mwongozi]

That type of 3D exploits an optical illusion. With the glasses on, one eye sees a darker image than the other eye, although both eyes are receiving a full colour image.

Because one eye is receiving less light, it takes longer for your brain to process the information coming from it. By the time it has, it is combined with the information being processed from the other eye. Because of the disparity in processing times, the two images combined are a short amount of time apart.

Thus can be exploited by rotating the camera around an object. By the time one eye has processed it's image, the camera has moved slightly, and the other eye processes its image quicker. This, the disparity in angles created a 3D image.

It only works when the camera is moving around an object in the right direction. As soon as it stops, the scene will look flat again, although you may think you are still perceiving depth because you brain remembers the previous depth information.

I've Done It

[ChrisMaple]

No, not 3D films, but I've made single stereoscopic images from multiple frames from television. A scene in which the camera has transverse motion is best; two frames can easily have the same vantage point spacing as a person's eyes. Motion of the actors works also (if everybody is moving in the same direction.) Good stereoscopic effects can even be achieved when the actor is rotating. The key is to get two different viewpoints for the same object. The effect is dramatically more vivid than anything my brain can devive from 2D television.

technology behind 2d-3d

[Anonymous Coward]

According to the patent: (USPTO#) 6,477,267

The displacement of the mesh sub-points may also be defined by a mathematical algorithm to thereby provide for automatic conversion of images. Further enhancements to the method could be to add shadow, blurring and motion interpolation data to the conversion data including force paralex information and field delay and direction for motion paralex delays.

Check out the patent for a full explanation of the technology.

An interesting form of 3D...

[cr0sh]

Some years ago, there was an episode of "That's Incredible" on which was displayed a system that showed 3D on regular TVs, without glasses, and the crazy thing was that you could close one eye, and still see the 3D effect! It was a box, that sat between the camera and the recording/broadcast equipment, and the resulting image was interesting, but it worked!

The image shown would "vibrate", it moved wonky, but there definitely was depth to the image. You could record the image, and play it back, and it was still there - a form of 3D that required no changes in broadcast or recording equipment, no glasses needed to view, and no special viewing system to watch - in short, it allowed 3D to be created by anyone, to be viewed by anyone (as long as they had one working eyeball!), on any standard video equipment. I have never seen this technology demonstrated anywhere else, nor did the company which presented its work (along with video clips that were fun to watch) go on to produce these boxes for sale - the technology and the company just seemed to "vanish" (is it any wonder?).

The closest I have been able to find about how this technology works can be seen here [well.com]. Please note that the site has "not safe for work" imagery on it...

This site's images, along with another poster's (below) comments about "temporal 3D" via running two movies out of sync, basically gives me a clue as to what they were originally doing:

I believe (now) that the box was somehow delaying the signal, every other frame, then interpolating those frames in/among the regular video frames and sending them down the wire. This isn't a very good explanation - basically, they were doing a combination of the temporal viewing with the "flicker GIF" of two stereo views (but without stereo, just time between the two frames) to generate the image. At the time, it must have been really expensive (for the RAM to buffer the image, etc) - although I wonder if they could have been de-interlacing frames and sending/reconstituting the frames by double-lacing the de-interlaced frames to make up the lost pixels, then showing each one (because each field of the frame would be out of sync by 1/15 second - maybe enough time to do the temporal 3D? - and it wouldn't require more than simple electronics rather than RAM buffering).

Aside from the flicker 3D images on the web (ie, those two different angle 3D animated GIF's like I noted above) - does anybody else remember seeing that episode of "That's Incredible", or anything else about the device? The episode was on in the mid-1980's or so...