BBC on 3d TV

Sorry for the lazy post - but I have been up to my eyes in prepping for
the Unconference for EverythingFuture.org.
I will get some more real life stuff down soon...In the meantime.

http://news.bbc.co.uk/1/hi/technology/10446419.stm
Use this link (above)for the full article.
FROM BBC.CO.UK
Quality warnings issued over 3DTV
Technology reporter, BBC News
Hollywood epic Clash of the Titans was painstakingly converted into 3D frame-by-frame but the image quality received mixed reviews. After years of trial and error, 3D finally hit the mainstream; at least as far as Hollywood was concerned.
Big box office hits like Toy Story 3, Avatar and Clash of the Titans suggest that cinema goers have an appetite for 3D and they're willing to pay a premium for it.
Now the TV industry is trying to catch up.
3DTV sets went on sale in UK department stores earlier this year, after sports network ESPN committed to broadcasting at least 25 of the World Cup games in 3D.
Despite poor first month sales figures - just 25,000 3DTV sets sold across Europe according to analysts GfK - industry pundits say 3D television is here to stay.
"3DTV is not going to have the same impact as the introduction of colour TV," declared Simon Murray, principal media analyst at Informa Telecoms & Media in a recent report.
"However, it is not a fad and it will prosper as a niche product," he added.
The average TV production has neither the time or the budget of a blockbuster movie. Filming in 3D requires both in large amounts, so the alternative is to film in traditional 2D and then convert it afterwards.
There are various ways of doing this.
All Samsung 3DTV's now come with TriDef as standard. If money is no object, then a proprietry post-production software system - developed by large 3D production houses - is the way many film studios are heading.
One firm, Prime Focus, converted Clash of the Titans into 3D for Warner Brothers in less than eight weeks. It was an expensive, labour intensive process, involving almost 1500 staff said the firm, and the results received mixed reviews.
Budget 3D
But there is one option that has no financial cost.
TV manufacturer Samsung recently announced that it will be including automatic conversion software - TriDef, developed by DDD - in all of its new 3DTV sets, enabling viewers to instantly "upscale" standard live television into a 3D experience at the push of a button.
TriDef is already embedded in Samsung mobile handsets and Acer laptops.
This type of software has alarmed the film industry, with claims that the automated result is inferior to filming with specialist 3D cameras or using an expensive post-production process.
At a conference in May, Hollwood film director James Cameron - the man behind the animated 3D epic Avatar - warned the result of cheap conversion were "eye strain and headaches".
Panasonic refused to include the software in its 3D sets over quality concerns.
"'We don't think it's right to confuse consumers this early on with second-rate conversion technology,' Fabrice Estornel, product manager at Panasonic TV, told website Home Cinema Choice (HCC).
Chris Yewdall, chief executive of DDD, said the company is not trying to compete with big budget solutions.
"3D is just like any other market - you have good, better and best," he told BBC News.
"We offer a very specific feature aimed at the consumer to get them comfortable with the fact that it is possible to watch 3D at home. There's a cost and quality level associated with that."
Comparing the product with surround sound, Mr Yewdall added that the DDD solution offers an unlimited amount of content as any live TV programme can be split into 3D without adding time and expense to the production cycle.
"When you look at Clash of the Titans , it had a very aggressive schedule and cost of several million dollars. The average consumer doesn't have that. You just want it to work at the press of a button," added Mr Yewdall.

Liverpool University Psychology Dept. and Vision Lab. trains 3d director in Stereography. Industry Insight Event planned in Manchester on Friday 18th

Courtesy of knowledge transfer funding from the North West Development Agency, Rob Black and Dr Bernard Harper have recently collaborated with RDA Innovation Award winner - Reg Sanders, classically trained (drama on single camera) ex-BBC director who has worked for among others BBC America, Discovery and the BBC.
Recently he began a project called Shoot3D.

The goal is to provide training and equipment for TV producers and directors wishing to capture 3D content for cinema, advertising and television.

Using their research in the Vision Laboratory in the School of Psychology as a base, Rob Black and Dr Bernard Harper provided media training tailored to the requirements of the client. This included distilling widely published research on human perception with studies conducted for the Independent Television Commission in Liverpool.

The data from a series of ITC funded experiments on Cyclopean Vision and
Orthostereoscopic Imaging from Presence journal and conference proceedings *
were the main resource used to guide the teaching and training. Over the
course of the stereographic training, the Reg was equipped with a solid grounding in the processes underlying visual perception. By greater understanding of how the
human brain processes visual information, stereographers are able to modify their
shooting and production grammar to produce more naturalistic and satisfying
3D content.

Much of the 3D media published at the moment is shot without knowledge of
basic principles, and as a result can produce easily avoidable perceptual
problems The key issues that have recurred throughout the history of 3D
imaging have been eyestrain and nausea often coupled to misperceptions of
size, shape, motion and colour. ** One theory is that visual disturbance
from poor stereography mimics the visual disturbances that accompany the
toxic effects of alcohol, poisons or of motions of the inner ear that lead
to seasickness. So our natural reaction to poor stereography is to feel
nausea and also eyestrain if the camera alignments force our eye muscles to
exert unnatural tension for extended periods of time.

Giving clients a grounding in visual perception enables professionals wishing to capture 3D content avoid many of the common 3D pitfalls and produce imagery that is easy to edit and view for greatly extended periods. Inappropriate camera alignments at the editing stage too can induce eyestrain through miniaturization, motion artifacts,
excessive parallax and incorrect object scaling so must be avoided.

Reg Sanders as a Stereographer with a strong visual background as a 2d drama director, along with Shoot 3D can offer assistance at every stage of the stereoscopic video production process. The steps include equipment specification, production
grammar advice, shot-by-shot analysis, on shoot shot, complete post
production services using industry-leading products such as Quantel Pablo
and Mistika and also projector calibration.

A 3d Industry Insight Workshop is being held at ProCam North, in Manchester on Friday, June 18th - featuring 3 x 2 hour sessions. For information contact reg@shoot3d.tv

A new convergence test.

Shot on a matched pair of JVC HD Handycams.
Non synced
Interlaced
at 70mm Inter Axial

http://www.youtube.com/watch?v=yeXVtM0Mw-Q

Ghost busting and Lanc synchronisation.

This was a sequence filmed under 'test' conditions with 2 HVR-A1 cameras.
They are not genlocked but ARE 'synched' using a Sony Lanc controller, they are
synchronized to within 5 milliseconds. We shot loads more - but on this shoot it wasn't very exciting,
so I have only posted the one clip. What I did learn from this shoot was about the concept of 'Ghost busting'
If you look in the background you can see the double imaging of the floor indicator above the lift doorway.
We converged on Rob (the guy in the black shirt) from about 3 metres - and the background has gone way beyond sensible convergence with the massive disparity that you can see. Well, this fact combined with the high contrast image creates 'ghosts' and there are some post production solutions that offer a 'ghost busting' service...
Youtube link here http://www.youtube.com/watch?v=ioXavo2hBNw

A 2d commercials director in the world of stereo 3d.

As my training continues I seem to understand less and less...
There is so much in Stereoscopy that seems to be counter intuitive when you have spent the last 20 years working in 2d. So far my position on the interocular is that you should approach each job differently.
I have read conflicting articles and been given conflicting advice, and I realise that the distance between the cameras is critical to the amount of '3dness' on the shot and how much of the shot will be usable or rather 'fuse-able'.
This is my take on I-O or if you prefer I-A so far... I'm not saying it is right - it just what I have learned so far.
Don't Shoot The Messenger.
The average distance between the human eyes is 64mm (2.5 inches for those that still use imperial). If you hold a coffee cup up in front of you at arms length and concentrate on the cup. You can see the front and a little of each side of the cup. If you could have a drink of some magic potion that would make you grow into a giant - say 12 feet tall - double your size in all dimensions you would hold that cup in front of you and be able to see that little bit more around each side of the cup. So if you shoot with your cameras say 130mm (5 inches) apart then you are replicating the view of that 'giant' person. Which means that the cup will look small, half the size. Or put another way, from the perspective of the cameras the cup has shrunk... So when you show these images in 3d the amount of 'depth' in the picture of the cup would indicate to your brain that you have turned into a giant, or more reasonably this is a smaller cup than 'normal' because you can see as much of the cup as you would have, had it been half the size! Stay with me, I did some tests and am happy to email them over... reg@shoot3d.tv. I bought 2 identical cups - one large and one small and took some very accurate still photos (based on the above) and compared the results - then I made 3d photos and showed them to a group of friends - there were no size cues in the pictures - just the stereo photo's of the cups. Without fail I could 'make' the big cup look smaller in 3d by using a larger interocular IO than is natural. So the 'Hyper Stereo' stuff that you may have read or come across is REAL. Then I found out that some of my sample viewers have enhanced 'stereo-acuity' (ability to judge / perceive 'depth' - relative size and geometry - based on interpretation of vertical disparities).
Now - where does this lead me. If you have seen any of the Sky TV 3d coverage of the UK football (Soccer) matches in 3d then those of you with high stereo-acuity may have found the 3dness a bit strange - and giving you the sensation that you were watching 'miniature' sized men running round a field! This is because some of the cameras had quite a large interocular. The view is generally that you can 'get away' with this in sport.

Here's a thought...
If you were shooting a commercial in stereo 3d, you have 2 guys, one with an aftershave product that is supposed to make him more attractive to the girls, then you might make him slightly 'larger' (hypostereo) by moving the cameras closer than the human IO (say by 10mm) and shoot the other guy in your commercial who is the 'loser' wearing the 'nobrand' aftershave - well, you could shoot him with a slightly larger I-O`(hyperstereo), again say 10mm - this might psychologically give an advantage to the 'hero'. I'll try it out and let you know if I get a script in for a 3d commercial.

What is Native Screen Pixel Parallax? NPP

What is Native Pixel Parallax?
Well, put simply - this is the concept that by basically using some maths, you can calculate 'what works' in pixels on your screen.
Explanation: When we shoot 3d with two cameras, it is the vertical disparities (the differences) between left and right camera that the brain 'sees' and then tries to 'fuse' the images and create an artificial state of mind that makes us 'see' depth or '3d'. So the distances between those cameras should be equivalent to or near to the natural 'interocular' which is about 65mm. (2 and a half inches in old money...). If the disparities are too much then it gives you a headache and makes watching a 3d sequence a very uncomfortable experience, in some cases if the disparities are too extreme it will mean there is no picture for the brain to fuse and it just shuts down any attempt to 'fuse' for depth.
Now here is the problem, if you shoot your 3d footage and for whatever reason you have had to shoot at a larger than 65mm interaxial, you may well end up with some stereo footage that 'works' on a small computer screen, but will not work on a large 3dhd plasma - or possibly, you may have a sequence that works on a 46" 3dhd plasma - and equally it won't work on a 30 foot cinema screen.
So some of the world's Stereographers have come up with a plan. They have come up with a numeric chart of 3d 'depth' that allows for average screen sizes and average distances from the screen. Obviously this isn't a perfect science, but it's a start... (Note to reader: It's quite hard 'blogging' about stereo 3d production, because not all the experts agree on certain principles... They all seem to have a different viewpoint to the same problems. So my position is try and be polite and take on board everyone's views and then when I am in the position of filming use the different advice as required....)
So here are some of the maths involved. You decide.
A 30 foot cinema screen = 360 inches (30 x 12inches) Human interocular = 2.5 inches
2.5 as a percentage of 360 = 0.7 percent - Now take that as a percentage of the resolution (2K)
then you end up with 14pixels. So any disparities larger than 14 pixels would represent an unnatural disparity which we have trouble fusing...(Based on 2k resolution on a 30 foot wide cinema screen)
Now try a 46" plasma. Width is less than 46" (The manufacturers get that measurement by measuring diagonally corner to corner of the screen) The width is about 40 inches so if we do the same thing and take
2.5" as a percent of 40" we end up with roughly 6 percent. If the Pixel resolution of the screen is 1920 x 1080
then 6 percent of 1920 should give you about 115 pixels. So theoretically you shouldn't go over 115 pixels on a 46" screen.
Simples?
When we look at 'Infinity' our eyes are parallel and anything we see across the depths will have 2.5" disparity.
The thing is when you 'scale' an image down (less than life size) to fit onto the 46" screen, then surely the disparities would be scaled down, or we are 'seeing' something totally unnatural?
If we only ever filmed Ortho stereoscopically (all measurements = real life) and then we only ever viewed our material back at life size on a large cinema screen - everything would be fine. A 6 foot man would be 6 foot tall and the disparities would be the same as real life - nothing over 2.5".
In reality we have to shoot for a range of different screen sizes - therein lie the problems. My conclusion is that whatever size screen you are shooting for always try and get the cameras as close to a natural condition as possible. If you are 'over' or 'under' the magic 2.5" then some post production can correct some a percentage of error.
Finally, you may deliberately shoot over or under for a special effect! But try not to blow the viewers mind! I am learning by trial and error. Everyday I shoot some more stuff and then work it out.

Motion artefacts / temporal occlusion 101 - JAN 2010

Yesterday, Simon Stappleton from Procam North and I, and Dr Bernie Harper from Liverpool University took two Sony XD cams off the shelf to take a look specifcally at motion issues as I had read some stuff on line about 3d blur and movement and how things can get tricky when shooting movement. I wanted to find out for myself what happened under different circumstances. During my 3d training I have come across loads of people in the film and TV bizz who are full of advice and are quite happy to tell you how it should be done... But then when I asked the questions, I found out that although everyone is ready with advice, NOT MANY have actually shot 3d!(I guess thats just a bit like life...)

We put the cameras onto the prototype Procam SbS Plate, a special plate that Jonathan Bolton at Procam North had engineered so that we could mount ANY camera Side by Side for trials. When you are training it's hard to get access to large mirror rigs, apart from the obvious reasons of cost, ie you need a team of Nasa trained scientists to program it, align it, and 'fly it' for you. Pretty much all of that technology is very expensive to hire, and because 3d is the latest 'in thing' particularly in London and the South east, access is very limited.

Training - is limited and expensive, but if you have the money then Pinewood and Shepperton run 3d training courses. Actually, on the subject of training I was very fortunate in being able to secure an 'Innovation Grant' from the RDA in our region, and that was used to get me superb access to some of the keenest and most intelligent minds on the planet who have been living and breathing 3d for 10 years or more. Now at last the technology is catching up...

For our motion test shoot we shot a variety of settings - interlaced, progressive, different frame rates. We filmed loads of cars driving by and I spent hours in Noise Industries Dashwood Stereo 3d Cinema Toolbox looking at disparities and basically trying to understand what worked, what looked good and what looked bad. Whether or not to track convergence on the passing cars or leave convergence set at an optimal distance. I will talk about convergence in a separate post because it is another of those 'sensitive'' areas in the 3d community, and the idea of this blog is not to cause stress and anger, but to raise the issues for discussion...
In terms of motion - Point One: I would always shoot progressively for motion. Full stop. end. new para.
Yes, you can play with frame rates and electronic shutter speeds - but you must shoot progressively because of how the frames with movement on have motion artefacts when shooting interlaced. You may have seen it yourself in 2d TV - that weird jagged edge where the scan lines have captured slightly different moments in time so a waving hand breaks up across scan lines. If you combine stereo shots with this effect you end up with a mess that the brain can not fuse, so no 3d. Always shoot motion progressively.
Point Two: Your cameras should be synced or 'gen locked' so that each frame is written and captured at the same time - that is to say the beginning of each frame must start at that 1/50th or 1/25th of the second (or 1/60th) and scan each line, both camera's perfectly in sync. Line 1 right eye camera, scans at the same velocity and the identical moment in time as Line 1- left eye camera and so on through the full frame scan.
So Progressive and genlocked / synced. We did try some early days tests with 2 HVR-A1's that were not 'genlockable' but were synced through a dual lanc within 100ms and on the whole the results were not bad. Though we weren't filming anything with much movement - just some stuff around 'sets' for a tv drama, kitchens, living room, that kind of well lit set up in a TV studio. We used a digislate to mark top and bottom of each take to see how the sync help up over time and it was reasonable and useable.
Definitions:Motion artefacts - weird bits of video that are created when the technolgy has to 'predict' what the movement in the frame is - or how the technology sometimes fails to capture the action properly.
Temporal Occlusion - when the differences between the two shots are caused by time, by the the way the 2 cameras have recorded or 'captured' everso slightly different moments in time, and so they don't work in 3d.
My final thoughts - you can shoot on interlaced formats - I use 2 little HD JVC's shooting AVCHDs or MPEG.TS for 'Pre Viz' and ' recce' . They are not locked, not synced and interlaced!
But they show me enough to see what works and what doesn't work...

A Christmas Carol - Not a good 3d experience DEC 2009

On one of those rare 'days off' moments - I decided to take the kids to see a 3d movie, kind of busmans holiday! Nevertheless we turned up at the local multiplex which had recently had its main screen converted to 3d.Infact, it was using the new Dolby 3d system which I had read was vastly superior to the Real D system, so I was keen to hand over my hard earned and get into the cinema... We took quite a load of kids with us, as our kids brought their friends who brought their friends! We ended up with 12 kids and 3 adults.

The first thing I noticed was that the glasses were duly handed out, beautiful shiny mirror look lenses and deep wrap around sides. I later discovered this is to maximise light the coming into each lens. During my stereography training I was told that the Dolby 3d system uses a combination of filtering to very finely control the colour information to each eye, so that although the channel separation is slightly less than the Real D polariser system the final result is better overall colour reproduction. I was looking forward to this screening, probably more than the dozen kids I was sat with!
I noticed there was a lot of activity pre-movie as the audience kept popping up and down to the attendant getting 'wipes' for their 3d glasses, getting them cleaned. I had to for some of our kids.
As the lights went down and the opening title sequence began I noticed I could see something in my lenses.
Again I cleaned them but it wasn't on the lens. What I could see was the reflection of my eyes and eyelids and lashes in the lens, as they were so shiny. Also the ambient light in the auditorium was very high, not helping the situation. In the UK they have strict regulations governing light levels in cinemas, and generally the cinemas have 3 levels. The walking in / walking out level well lit 'house lights' - then the next level of dimming is the spot reel or the adverts before the main presentation, I assume they have that lighting so that we can juggle our popcorn and kids, and get settled for the main movie. Finally, darkness for the main feature.
However, in my cinema trip, the lights never dimmed down to full darkness for the movie. You could extend your hand out and clearly see a shadow cast on the floor from the lights above. This caused problems with the 3d viewing because the images I could see through my lenses were overlaid with reflections, also the movie itself had a fairly 'dark' feel to it (being set in Winter in Dickensian London).
And therein spoilt what should have been a great 3d experience. The trouble is these multiscreen movies have very few people involved in their running. It's all done by computers... as they say. A Christmas Carol started and I went off like Scrooge trying to track down the projectionist. I went to the front of house area, to find a couple of students selling popcorn and coffee, they told me the projectionist had 'gone home' and would be back later to turn everything off... I asked for the manager, there was none. So the next day I'm on the internet and phone trying to find someone from the Vue Cinema chain to talk to... No luck. I sent an email to their customer services team who eventually reply with some brush off corporate letter about local planning regulations on light levels in public buildings... And didn't take on any of my points about 3d needing more careful light control than 2d so that the viewer can properly extract disparities and depth information for proper 3d viewing. I think they were worried about me trying to get my 50 quid back from them for the trip, but I made it clear I just wanted some technical dialogue on their procedures and spec for 3d cinemas... Bah humbug!

Reg Sanders Contact details.

Reg Sanders
07764 354 984
email: reg@shoot3d.tv

UK based Writer / Director
3d Stereographer
www.shoot3d.tv
http://shoot3dtv.blogspot.com/
LinkedIn: http://uk.linkedin.com/in/regsanders
UK Commercial representation: www.mobfilm.com
Online showreel: www.youtube.com/hungrywolffilms
Online Viral: www.youtube.com/theviraldirector

Google: Reg Sanders
Skype: Reg.Sanders