Monthly Archives: June 2011

Consumer HD, D-cinema, Post Production

Subtitles in Stereoscopic 3D

Must float on top.

Stereoscopic 3D subtitles are a bit more complex than their 2D counterpart.

The apparent depth of any 3D content can range from infinity behind the screen all the way to the theater space.

Conventional subtitles typically use single color font with an optional drop shadow or an extrusion. Since such subtitles have no stereoscopic parallax their apparent 3D depth places them on the screen plane.

This would be fine if all other content were placed behind the screen and none projected inside the theater space, but this is rarely the case. Placing subtitles on the screen plane while other content is seen in front of the subtitles creates a powerful visual conflict.

To address this issue, S3D subtitles’ depth must change from shot to shot, and on a rare occasion may even change dynamically to accommodate for depth changes in a shot. The challenge is to avoid depth conflicts while keeping the text legible.

3D Blu-ray supports subtitle parallax rendering using depth information contained in the metadata. Theatrical digital projectors currently do not support this feature and require composited (burned-in) subtitles for each foreign language DCP.

Technical Info

Stereoscopic Zone of Comfort

Where eyes don’t hurt.

Stereoscopic 3D zone of comfort is the range of positive and negative parallax (apparent depth behind and in front of the screen) in which discoupling of accommodation and vergence can be achieved without discomfort.

Excessive parallax between the left and the right eye view on the screen causes eyes to converge or diverge too much which can cause eye strain. The actual boundaries of zone of comfort vary slightly from person to person.

Post Production

Stereoscopic Inpainting

A step in 3D conversion.

During 2D to stereoscopic 3D conversion objects in the scene are cut out, placed in depth and given volume. Because of the parallax disparity between the left eye and the newly created right eye, areas that are in real life occluded by foreground objects are now showing as holes in the middle ground and background.

Skilled artists paint these areas in which is referred to as “inpainting.” There are automatic and semi-automatic tools that aid the process, but manual approach works the best for high quality conversion.

Post Production, Technical Info

Dense Disparity Map (Depth Map)

It maps where objects belong in stereoscopic 3D space.

Dense disparity maps (depth maps) are grayscale images that determine the Z depth placement of each pixel in a 2D image. The concept is equivalent to Z buffer maps in CGI animation. Objects closer to camera are rendered closer to 100% white and infinity is rendered as pure black.

A depth map can be used in conjunction with complimentary full color 2D image for a variety of stereoscopic post production processes. For example, a depth map can be used to create additional multiscopic views.

Technical Info

Holographic Television

Look ma, no glasses!

Holographic TV is under development at MIT and several other research labs worldwide. The goal of holographic TV is to provide depth-rich viewing experience that parallels real world experience without need for special glasses.

In other words, looking at a hologram is physically and physiologically equivalent to looking at a real life scene. Holograms can’t, at least not yet, make objects float in space like seen in SciFi movies. However, they can render a fully accurate sense of 3D depth that extends behind and in front of the screen.

Equipment

Autostereoscopic Displays

3D without glasses.

Autosterescopic video displays do not require viewers to wear special eye wear. The challenge in designing autostereoscopic displays is to provide discreet left and right images to appropriate eyes while avoiding cross talk.

Some technologies like holographic TV are promising but are only at experimental stage. Lenticular displays are commercially available at this time but do not provide the full depth quality as their eye wear based counterparts.

Technical Info

Vergence/Accommodation Conflict

Can make head hurt.

Vergence/accommodation conflict is often quoted as one of the contributors to why some people experience discomfort when watching stereoscopic 3D (S3D) films or TV.

When we look at a nearby object in real life, we focus our eyes at it (accommodation) and we rotate our eyeballs inward to allow each eye’s gaze to cross at the object (vergence). Distinct groups of muscles are in charge of these two operations and typically work in unison.

However, when we watch an S3D film, we focus at the screen plane, but we may converge our eyes in front of or behind the screen. This decoupling of vergence and accomodation can contribute to eye muscle fatigue.

ARTICLES

Interview With Stereographer Daniele Siragusano

The 2nd SMPTE Annual International Conference on Sterescopic 3D for Media and Entertainment took place in New York on 21st and 22nd of June 2011.

Over twenty researchers and engineers presented papers on topics ranging from stereoscopic 3D viewer comfort to advancements in holographic 3D.

Igor Ridanovic spoke to German stereographer Daniele Siragusano who presented his research on stereoscopic volume deformations at the SMPTE conference.

IR: Please tell me about the facility you work for and your recent stereoscopic 3D projects.

Daniele Siragusano at SMPTE conferenceDS: I work for Germany’s largest post production company CinePostproduction. Founded 100 years ago as Geyer film lab in Berlin today we offer the complete range of film processing, digital post production, workflows and on set services for digital and stereoscopic cinematography including DCPs and prints for theatrical release in Munich, Berlin, Hamburg and Cologne.

We did the post production for the first European animated stereoscopic 3D film Animals United. We did Stereo-sweetening for The Darkest Hour and the post production for a converted movie called Hybrid. Currently we are involved with the S3D sweetening of Vicky and the Treasure of Goods and we are doing sound mixing of The Three Musketeers at our sound studios in Munich.

We’ve done a lot of commercials and shorts as well although we’re not a commercial house. But production companies come to our facilities because we know how to do 3D. We’ve done shows for German car manufactures and two or three projects with universities.

IR: Your approach to stereoscopic post production combines rigorous testing which is not unlike that of the researchers who presented papers at the SMPTE stereoscopic conference, but you also have the advantage of working on live action material. Are your research findings ever at odds with what you see on the screen?

DS: Yeah, the good thing about my position is that I can do all the research and see the final product on the screen, and since stereoscopic perception is a cognitive process, you can do all the math and it will only explain one part of the whole story. But there are lots of cognitive processes as well and you can not really put them into numbers. It’s really important to understand both the theory and the real life results.

IR: Perhaps it’s fair to say the scientists who study the stereoscopic 3D are at disadvantage because they don’t get to work with live action material on large screens?

DS: One part of the scientific process is to isolate and explore one special subject and ignore the sum of all parts. When it all comes together in one image, one frame, many factors are added together. It’s important to have the research but you also have to experience the S3D.

IR: How much R&D do you put into each film project?

DS: Quite a lot. For every big show there is a lot of R&D going on. There is always something new.

For example, we had a show in which the director wanted to add grain because there was a film within the film plot. We started to experiment with grain in stereoscopic 3D. We put different grain in each eye but this was unfusable by the viewer. Then we added the same grain to both eyes, but this looked like a wall of grain in the 3D space. Finally, we converted the grain so it had the same depth as the object it occluded. The visual effect was that the surfaces were grainy.

We’re also digging into new ways of modifying stereoscopic characteristics and changing the depth of individual objects in the scene in order to support the storytelling.

IR: In a recent film you manipulated the vertical parallax in order to induce unpleasant sensation in the viewer. Do film directors typically understand the new tools that are at their disposal or do you find yourself in the position to educate them?

DS: We definitely consult with directors about the possibilities. For a lot of directors I work with it’s the first time they do stereo and they appreciate any help they can get. For the post production side of things we’re getting involved much, much earlier, even before the storyboarding. They come to us at the early stage of the production so we can show them what flexibility they have in post and what they should get right on set—what we can not fix in post.

IR: What’s the situation in EU and in Germany in particular regarding stereoscopic training for film and TV industry professionals?

DS: I am teaching a course at the University of Television and Film Munich and there are stereoscopic consultants who offer courses, but there could be more, it’s just the beginning. I work for several universities as a lecturer. They are starting doing R&D on the university side, but I think it will take two to three years until they really get this deep understanding of how to deal with real life S3D problems. But we get some educated people from the universities, who have the basic understanding of stereo.