How to make proper 3D content. - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 56 Old 11-08-2010, 05:01 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Short description of what filming is.

The camera captures light from what's being filmed.

Short description of capturing the light.

The light fits into the area the camera see's it, and not more or less area than that.
The size of the area the light is filmed in changes the size of the light being captured.

Short description of seeing the captured light.

To see the light of what's being filmed you need to see from the area the camera captures it.
It's possible to see the captured light from a different area it was captured in, and when this happens you don't see the captured light exactly as the camera saw it.

The viewer should see the filmed light at the intended size.
If you see the light of what's being filmed from a different area than where the camera was when it captured the light, your seeing the size of the captured light differently than the camera saw it, and this makes the size you see different than the size the camera saw.

So maintaining the area is important when watching the captured light, the quality of the area your watching the film in is judged by whether or not the captured light being seen is the same size the camera filmed it at.
So there is a new term I'll coin - "quality area" .

Short description of Bad Filming - Zoom.

When watching the captured light in a quality area, what's seen is in a naturally proportioned size - Faces don't seem too large that it's strange to see them.

There's a problem when the area is quality and the faces are too large or small when first seen: this creates a problem with motion as well, when they move they seem to move exaggeratedly.
To fix this you would see the persons face at a normal proportion for the area it was filmed in, and then dolly the camera to zoom in or out and so change the area to a larger one or smaller one.

The same thing for the environments foreground and background, it exists in the area the light is captured in.

Short description of Good Filming.

The person or thing is in the area your filming.

The shot begins with the area. The area holds the background and foreground in context of their natural proportion, or you look at the area and everything seems neither too large or small.
The character is then shown within the area.
If the area is not holding the background and foreground in proportion, when the character is first seen he or she is too large or small for what seems natural and the speed the character moves at is also off.

Then with the background and foreground set up so when the character is first shown he or she is in natural proportion, if the person wanders out around the area he or she will stay in proportion: the camera may stay stationary, dolly up to them, or turn following them from one stationary spot.

Think of the area as a web-page and the character as font size. The size of the web-page when you see the font is neither too large or small for what seems natural.
Then the font changes size on the web-page so it becomes either larger or smaller. Seeing the font at a exaggerated size when first seeing the web-page is not seeing it as a natural web-page.
Different text is different size, but there's stationary or regular font size.

Short description of Good Filming using technical details specified in terms of stereoscopic camera geometry as an equivalent of human vision system.

The simple rules are.

• Fixed HFOV field of view of 40 deg, target screen size 48 inch (diagonal) screen.

• The stereo window of the camera is the same size as the target screen.

• The view angle of the camera is the same as the viewer view angle.

• No zoom. Best focus and maximum DOF as permitted by the lens system.

• To project such content on different size of the screen one must simply float the window in or out to transform the stereo-window into its original camera position.

Audio of the text

____________________________________________________________ ____

Mathew Orman and Joe Clark, you two can discusss this threads first post to agree or disagree on points on how a 3D stereoscopic film should be made.

I know people in the movie business read this forum so I thought it would be a good idea to post this topic here. Me and Mr.Orman discussed this over at 3dvision-blog.com already.

__________________________

The above was for Mr.Orman to discuss the making of 3D with Mr.Bloggs, but sofar nobody has wanted to talk to him about making 3D, until now when I read Walter Murch talk about 3D: link

So after reading Walters letter to Roger Ebert I sat down and wrote the following below (I hope Mr.Orman doesn't get upset that somebody else's ideas are presented here besides his own):

3D is the perception of real distance when looking at the screen, 2D accomplished this.
"Once the x- and y-axis are specified, they determine the line along which the z-axis should lie" - wikipedia, Cartesian coordinate system.
This means once you've established the 2D x and y coordinates the z axis may exist.
2D video x and y axis brings a perception that the z axis exists, and this affiliation is seen even though there is only x and y axis.

When you see the z axis but it removes you from the story's immersion, it's because it's showing that the z axis is seen.

Test to show you 3D.
1.) Hold two fingers in front of your left or right eye and form a plus sign +, now you see how to hold your left and right hands, thumbs facing your face, in front of your left or right eye. Remember to keep both eyes open even though your holding your hands in front of only one eye.
2.) Open your computer and Notepad, make notepad full screen and then look at the white screen made by Notepad with your hands in a plus sign +, as described in step one.
3.) Now to see the 3D effect poke your thumb out towards your eye, so the thumb is pointing right in front of your eye, this is the z axis.
Now try without your thumb pointing in front of your eye, this is the x and y axis.

If your trying to see what's going on in Notepad, when your thumb pokes out it takes your attention away from the Notepad screen, doesn't it?

Presbyopia: is a age related eye disorder. What Presbyopia does to the eyes is it progressively diminishes the eyes ability to focus on near objects.

The problem of 3D TV, is that the audience must focus accommodation their eyes at the plane of the screen -- say it is 80 feet away. This is constant no matter what.
But their eyes must converge at perhaps 10 feet away, then 60 feet, then 120 feet, and so on, depending on what the illusion is.

Most people over the age of 50 lose the ability to link convergence and accommodation cues, due to Presbyopia.
All living things with eyes have always focussed and converged at the same point.

When the convergence changes, the eyes focus there. With Presbyopia the convergence changes, the convergence will be further in, but the focus cannot follow it.
This makes the mind concentrate on the convergence and not on the focus.
How this fits into films is if the mind concentrates on convergence and not focus the person identifies the video as 3D, or immersive, but if the mind focuses on the convergence and focus at the same time then there is zero parallax and so no 3D effect or immersion from positive and negative parallax.
The young eyes can see convergence and focus at close range, but the old eyes cannot because of Presbyopia, so the old eyes have the effect that 3D offers, and not 2D.

The Brechtian trick described by Walter Murch.
"Brecht created an influential theory of theatre, the epic theatre, wherein a play should not cause the spectator to emotionally identify with the action before him or her, but should instead provoke rational self-reflection and a critical view of the actions on the stage.

For this purpose, Brecht employed the use of techniques that remind the spectator that the play is a representation of reality and not reality itself [...]. Such techniques included the direct address by actors to the audience, exaggerated, unnatural stage lighting, the use of song, and explanatory placards." - wikipedia.

To fix this Brechtian effect the z axis brings to video, you wave the z axis back and forth, gently: this waving motion breaks the focus on the z axis and then the focus goes back to the x and y axis presentation.
It matters if you wave the z axis in a 360 degree circle as compared to moving in either the x or y directions.
When you wave the z axis in a circle the focus remains on the z axis, but wave the z axis in either the x or y axis direction and the focus moves away from the z axis.
To test this, do the 3D test described before, but this time wave the hands around to the left and right, then up and down, then in a circle -You will see the Notepad screen becomes more or less visible.

Back to discussing 3D displays now, on a volumetric 3D video, the z axis when shown can be put into context of the x and y axis, because the eyes look at the z axis, then they look at the x and y axis.
To move the eyes from the z to the x and y axis, and keep focus, the eyes do not move in a circular direction - like your rolling your eyeballs in your head.

So, to define what 3D effect is : The mind concentrates on the convergence, then back to the focus and convergence, because concentrating on only the convergence and not convergence and focus may create the Brechtian distraction: not being able to focus on the convergence and accomodation seen in 2D is not a good thing, it's old man eyes.

Wikipedia says the root word of Presbyopia is Old man + sightedness, it sets in at around 40+ years of age: some people think 3D is Awesome, so old man eyes may make the world look Awesome, making the concerns 2D brings moot.
 
To tell a story, contrast zero parallax with positive and negative parallax: the main story running time is made up of over the top Brechtian theater - when the plot driven part of the story is shown, it gives feedback to this. Seeing convergence and focus in a 3D film is seeing 2D, or zero parallax - there is no Brechtian effect from the z axis, it's all story driven.
 
The story is told in z axis 3D, then 2D shows this story in a new light. In subliminal marketing, it may be that one characteristic of the topic subject is bolded and another characteristic is muted: the bolded characteristic is in 2D and the muted characteristic is in z axis 3D. 3D (Brechtian effect), gives a critical view of what happens in 2D.

Audio version


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
Sponsored Links
Advertisement
 
post #2 of 56 Old 11-08-2010, 07:53 PM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Great!
Good start. It is not difficult to setup distortion free system on both capture and projection ends.
The most difficult part is camera action in story telling.

In stereoscopic display systems the 3D volume of a scene is divided into two parts by a screen plane.
The part that is between our eyes and screen overlaps two worlds, the real and virtual which was captured by stereoscopic camera. The second part which extends from screen to infinity is totally virtual.
What it means is that in the front part viewer can inter-react with virtual world by using his/her heads or just about any objects that are visible.
It is an extremely useful phenomena and it has being utilized in systems which are called augmented reality.
However there is a problem when it comes to presenting
3D movies. The objects of a scene that are on negative side of stereo window become sliced and look like floating in front of the screen. This is unnatural and seldom observed in reality. So here special filming techniques must be used to make sure that all scene objects stay behind the screen in virtual portion of 3D scene.

The simplest way is to make a frame which is of size that is exactly the same as the target stereoscopic window.
The more expensive method would employ 3D volume reconstruction in real time and visual stereo window clipping of objects that extend into negative space in-front of the screen.

Mathew Orman

http://www.*******************-usa.com/
icester is offline  
post #3 of 56 Old 11-08-2010, 08:33 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
May I reword what you said so it is easier to read? Please look at the text below and verify I understood you correctly.

_______________________________________

The active polarized glasses divide the screen into two parts - using a 2D plane twice.
The second 2D plane overlaps the first, creating a stereoscopic effect: the actual 2D frames and the virtual world sensed while viewing stereoscopic content.

The problem is when showing negative parallax, they look like they are floating slices in front of the screen.
So here special filming techniques must be used to make sure that all scene objects stay behind the screen in virtual portion of 3D scene.

The simplest way is to have two 2D frames form a single stereoscopic perimeter measurement and fit the TV shape to this stereoscopic frame.
There would be black borders on the left and right side depending on which 2D picture was being shown.

The more expensive way would be to cut the two 2D frames so they fit into a standard monitor's aspect ratio.
So the TV screen shows the picture but either the left or right side has black clipping, the clipped side then shows video and no clipping and the clipping goes to the other side of the screen.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #4 of 56 Old 11-08-2010, 08:56 PM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

May I reword what you said so it is easier to read? Please look at the text below and verify I understood you correctly.

_______________________________________

The active polarized glasses divide the screen into two parts - using a 2D plane twice.
The second 2D plane overlaps the first, creating a stereoscopic effect: the actual 2D frames and the virtual world sensed while viewing stereoscopic content.

The problem is when showing negative parallax, they look like they are floating slices in front of the screen.
So here special filming techniques must be used to make sure that all scene objects stay behind the screen in virtual portion of 3D scene.

The simplest way is to two 2D frames form a single stereoscopic perimeter measurement and fit the TV shape to this stereoscopic frame.

The more expensive way would be to cut the two 2D frames so they fit into a standard monitor's aspect ratio.
So the TV screen shows the picture but eith the left or right side has black clipping, the clipped side then shows video and no clipping and the clipping goes to the other side of the screen.

Not correct.
You need to learn about what is called view thrustum which extends from an eye to screen and in 3D system beyond the screen to infinity.
In stereo projection system both parts can contain 3D objects. But only the front part can contain both physical and virtual objects.
I has nothing to do with shutter glasses.
Maybe a 3D scene in anaglyph will help you grasp the concept. There is plenty of drawings, just search for 'stereoscopic camera frustum'.

Here is an example.



And here is an example with both parts visible in anaglyph.
Notice that you can insert you finger into the object that is floating in the front part of the frustum.



Mathew Orman
icester is offline  
post #5 of 56 Old 11-08-2010, 10:09 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
The toe In method is both left eye and right eye cameras are pointed at a single focal point.

Off axis method is separating the left right eye cameras.
1. The cameras should be separated 1/20 the distance to the screen.
2. Negative parallax should not exceed eye separation.

link

Off axis has extended frustum. Mr.Orman, are you suggesting that the extended frustum should not be cut off when viewing the screen?


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #6 of 56 Old 11-08-2010, 11:59 PM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

The toe In method is both left eye and right eye cameras are pointed at a single focal point.

Off axis method is separating the left right eye cameras.
1. The cameras should be separated 1/20 the distance to the screen.
2. Negative parallax should not exceed eye separation.

link

Off axis has extended frustum. Mr.Orman, are you suggesting that the extended frustum should not be cut off when viewing the screen?

No,
that is 3D gimmick rules of how to create totally distorted content.

The camera separation must always be and remain 65 mm.

The only deference between to-in and off-axis is the keystone distortion.

We have established the requirements of realistic stereoscopic camera and projection system and now defining 3D filming techniques that eliminate unnatural effects which due to scene's object slicing by front of stereoscopic view frustum.

Mathew Orman
icester is offline  
post #7 of 56 Old 11-09-2010, 01:06 AM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I edited the five rules in the first post. May I ask you to look at them and correct them if their wrong, Mr.Orman?


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #8 of 56 Old 11-09-2010, 02:24 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

I edited the five rules in the first post. May I ask you to look at them and correct them if their wrong, Mr.Orman?

It is your post and I cannot edit anything.
It was correct the first time and now is completely wrong.

The target screen size and HFOV (horizontal field of view) defines the viewer distance which is the same as the original camera to stereo window distance.
The stereo window of the camera is the same size as the target screen and the view angle of the camera is the same as the viewer view angle.
There are no other rules.

Mathew Orman
icester is offline  
post #9 of 56 Old 11-09-2010, 12:07 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Thank you for correcting me on my mistake. I have re-edited the first post.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #10 of 56 Old 11-09-2010, 07:32 PM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post
Thank you for correcting me on my mistake. I have re-edited the first post.
Still wrong.
You provide links to geometry that is parallax driven and not
correct for realistic rendering of stereoscopic content.

The distance between lenses is 65 mm and not 1/of what ever.

If you have your own theories you want to promote then please exclude me from.

I do not propagate gimmick stereoscopic content and or systems.

Mathew Orman
icester is offline  
post #11 of 56 Old 11-09-2010, 08:15 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by icester View Post
If you have your own theories you want to promote then please exclude me from.

Mathew Orman
I'm sorry for having included my own theories. I have edited the first post removing the content I thought up.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #12 of 56 Old 11-09-2010, 08:24 PM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post
I'm sorry for having included my own theories. I have edited the first post removing the content I thought up.
Great!
Now we can get on and create the first fully realistic and undistorted stereoscopic video content for the 48 inch screen and other popular sizes that viewers might request.

Mathew Orman
icester is offline  
post #13 of 56 Old 11-09-2010, 09:36 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I read online that James Cameron's Fusion System is Toe in Based.
With Toe In "The objects of a scene that are on negative side of stereo window become sliced and look like floating in front of the screen.". (If I'm quoting you correctly I hope I am.)

But with Off Axis this floating is not there? I am wondering what this means though "The simplest way is to make a frame which is of size that is exactly the same as the target stereoscopic window."

I think you said a filming technique that when used with Off Axis would make the floating go away. What would that filming technique be, I could not grasp what you meant in the sentence I quoted.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #14 of 56 Old 11-09-2010, 10:31 PM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

I read online that James Cameron's Fusion System is Toe in Based.
With Toe In "The objects of a scene that are on negative side of stereo window become sliced and look like floating in front of the screen.". (If I'm quoting you correctly I hope I am.)

But with Off Axis this floating is not there? I am wondering what this means though "The simplest way is to make a frame which is of size that is exactly the same as the target stereoscopic window."

I think you said a filming technique that when used with Off Axis would make the floating go away. What would that filming technique be, I could not grasp what you meant in the sentence I quoted.

Again toe-in and off-axis differs only by keystone distortions.
That has nothing to do with object being sliced by stereo window.

To prevent object from being sliced simply back off the camera or use a lightweight metal frame which would be place where the real stereo window is.
It is like a guard rail that prevent objects or actor from getting to close to the camera passing the stereo window distance.

James Cameron's toe-in camera wasn't ever preset to stereo window equal to cinema screen size and worst of all the camera varied IO together with focus and convergence
coupled just to satisfy minimum and maximum parallax value. Which resulted in totally confusing experience.
Imagine an actor growing in size and getting squashed in depth at the same time during just a second or two of action. Human brain cannot be fulled when it comes to stereo vision and perception of size and distance.

Mathew Orman
icester is offline  
post #15 of 56 Old 11-09-2010, 10:42 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Thank you for explaining that Mr.Orman.
I have no other questions.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #16 of 56 Old 11-09-2010, 11:32 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
I added these quotes to the first post Mr.Orman. That way the important info is easy to find instead of having to go through all the posts we made where I made bad info.

________________________________

"The objects of a scene that are on negative side of stereo window become sliced and look like floating in front of the screen.".

"To prevent object from being sliced simply back off the camera or use a lightweight metal frame which would be place where the real stereo window is.
It is like a guard rail that prevent objects or actor from getting to close to the camera passing the stereo window distance.

James Cameron's toe-in camera wasn't ever preset to stereo window equal to cinema screen size and worst of all the camera varied IO together with focus and convergence
coupled just to satisfy minimum and maximum parallax value. Which resulted in totally confusing experience.
Imagine an actor growing in size and getting squashed in depth at the same time during just a second or two of action. Human brain cannot be fulled when it comes to stereo vision and perception of size and distance.

Mathew Orman"


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #17 of 56 Old 11-10-2010, 05:03 AM
Super Moderator
 
markrubin's Avatar
 
Join Date: May 2001
Location: Jersey Shore
Posts: 22,673
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 25 Post(s)
Liked: 256
sticky

Please take the high road in every post
Please do not quote or respond to problematic posts: report them to mods to handle
Link to sponsors
good to be back to vBulletin
markrubin is offline  
post #18 of 56 Old 11-17-2010, 03:51 AM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by Joseph Clark View Post

Mathew,

I bow to your obviously much greater knowledge of 3D. In all seriousness, I'm here to learn, and I appreciate the years of experience that people such as yourself bring to these discussions. But, I'll ask you here what I've asked you in other threads - what specifically is it that you want 3D filmmakers to do differently? In terms of 3D cameras and shooting techniques, are you saying that there is only one right way to shoot 3D? Are you saying that because he chooses to shoot 3D "unrealistically" that James Cameron's movies are bad? When you say Avatar is the worst 3D movie ever made, the generalization seems extraordinary.

When you ask the questions, "Who wants to see an actor growing in size and getting flattened in just a second or two of action time.
Who wants to see nothing but gigantically scaled version of actors and robots that move as if they ware made of Styrofoam." The answer appears to be: if the box office for Avatar is any indicator, over 2 billion dollars worth of them want to see that. I did - four times. Now, I'm not trying to say that because it's popular 3D, it's right. But you do seem to be saying that because it's not realistic 3D, it's wrong. I'd respond to that by saying that little about filmmaking has ever been realistic, even in 2D. As I said in a recent response to one of your posts, 2D filmmaking is full of unrealistic two dimensional distortions, including those created by wide angle, telephoto and zoom lenses, vertigo shots, camera dollies, sudden high angle to low angle transitions, etc, etc, etc, etc. These things have been used for over a century to engage and entertain people. Are you condemning such 2D lenses and shooting techniques because they're not realistic? Are you asking 2D filmmakers to stop using them? Are you asking 3D filmmakers to stop using unrealistic 3D cameras and shooting techniques and only shoot 3D the way you think it should be shot?

Maybe I'm simply misunderstanding what it is you're saying, and asking. Obviously, you have a lot of experience, and you spend a great deal of time thinking about these issues. But there seems to be a disconnect between your vision of what 3D filmmaking should be, the reality of what 3D filmmakers are doing and what people like seeing in their 3D entertainment. I know I've asked this question before, and I know it sounds like I'm calling you Quixotic, but I'm really curious about what your goals are. Please have patience with me. I honestly do want to understand your position better than I do.

link

After this question from Mr Joe Clark, Mr.Orman asked the question to be put here so they could discuss the topic in this thread.

I will try to answer Mr.Clark and wait to see if Mr.Orman replies and corrects me.

__________________________

Mr.Orman is concerned with the negative parallax being too close to the camera, so he suggests that a light weight bracket be put on the camera to keep the actor at a safe distance. This is to keep the flat surface artifact at a minimum.

The idea of 3D is to have two parts at a distance they don't conflict with the stereoscopic cue for depth perception.

These two parts are:
1.) the foreground, it hold the interactive thing or person and the environment being interacted with.
2.) the background, it hold the environment being interacted with.

The camera needs the environment to hold the person interacting with it so the two parts I numbered remain uncompromised.
It is not hard to do, the thing is the environment is in two parts and it holds one part the thing that interacts with it.

__________________________

Mainly films use this method I'm going to mention, I think it's safe to say as I have watched a fair number of movies, but I'm no director.

1.) The person or thing is seen in one environment: introduction of the character or scene: it may only show a close up.
2.) The environment has a blur to it due to fast movement, the background is moving fast.

The error Mr.Orman is seeing (I think), is due to the fact that :

1 is not at the proper distance so the viewer can watch the environment being interacted with in a background and foreground so it has proper stereoscopic cue,

2, I'm assuming involves a closeup, is without premise of it's stereoscopic position when it should first show how the fast moving scene fits in with the stereoscopic environment then move into the scene via dolly or best focus.

__________________________

The environment is a container and the person or thing interacting is in this container, the thing is to keep these in perspective.

If you show only the fast environment does it have the cues of a background and foreground?
If you show the person or thing that interacts do they exist realistically in the environment that contains them?


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #19 of 56 Old 11-17-2010, 05:41 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

link

After this question from Mr Joe Clark, Mr.Orman asked the question to be put here so they could discuss the topic in this thread.

I will try to answer Mr.Clark and wait to see if Mr.Orman replies and corrects me.

__________________________

Mr.Orman is concerned with the negative parallax being too close to the camera, so he suggests that a light weight bracket be put on the camera to keep the actor at a safe distance. This is to keep the flat surface artifact at a minimum.

The idea of 3D is to have two parts at a distance they don't conflict with the stereoscopic cue for depth perception.

These two parts are:
1.) the foreground, it hold the interactive thing or person and the environment being interacted with.
2.) the background, it hold the environment being interacted with.

The camera needs the environment to hold the person interacting with it so the two parts I numbered remain uncompromised.
It is not hard to do, the thing is the environment is in two parts and it holds one part the thing that interacts with it.

__________________________

Mainly films use this method I'm going to mention, I think it's safe to say as I have watched a fair number of movies, but I'm no director.

1.) The person or thing is seen in one environment: introduction of the character or scene: it may only show a close up.
2.) The environment has a blur to it due to fast movement, the background is moving fast.

The error Mr.Orman is seeing (I think), is due to the fact that :

1 is not at the proper distance so the viewer can watch the environment being interacted with in a background and foreground so it has proper stereoscopic cue,

2, I'm assuming involves a closeup, is without premise of it's stereoscopic position when it should first show how the fast moving scene fits in with the stereoscopic environment then move into the scene via dolly or best focus.

__________________________

The environment is a container and the person or thing interacting is in this container, the thing is to keep these in perspective.

If you show only the fast environment does it have the cues of a background and foreground?
If you show the person or thing that interacts do they exist realistically in the environment that contains them?

Well,

you are missing on the big one and that is human brains ability to estimate physical size of objects and their associated motion inertia of mass. If natural stereoscopic vision geometry is altered to satisfy movie theater screen parallax limit the brain of a viewer gets total confusion as
it detects such changes in natural proportion of size and inertia. Also it is unpleasant and unnatural to view an actor who's size is several times different then actual size.
Most of all telephoto shots of actor faces always makes it look distorted and an unpleasantly deprived of natural perspective.

They will never change those techniques as they are convenient and low cost both for the directors and actors.

The argument that Billion dolor profit is a prove that Cameron's gimmick is what everyone should be using to create movies is false since it is the first 3D movie after
several years of absence and the ticket prices ware extraordinary. Since TITANIC many new theater ware built around the World and US$ value had declined significantly.

In any invent viewer are starting to recognize the difference between gimmick and realistic 3D.
An example is the movie called:
Grand Canyon Adventure: River at Risk 3D

http://www.blu-ray.com/movies/Grand-...Blu-ray/13544/

The movie has many scenes with geometry is compatible with several home theater screen sizes.

It is a breath taking and extremal immersible experience.
With some scenes on can experience depth extending to infinity.

Sooner or latter there will be more movies like that and
movie makers will have to choose between satisfying their own conveniences and satisfying viewer experience quality.

Mathew Orman
icester is offline  
post #20 of 56 Old 12-09-2010, 10:10 PM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
What the rules mean, in a short crude description. Is the distance you see the screen, you need a certain amount of parallax.
This is so you see the picture in 3D.

If you have parallax but not enough for the distance the picture is seen, then you don't see 3D and it makes you feel odd.

So when you make the video content, you make the parallax fit a TV size, so
1.) you actually have the TV that is that size,
2.) you make the content with different parallax,
3.) you choose the parallax you feel comfortable with for that TV size viewed at that distance.
4.) you create content with that parallax.
5.) fast moving content is harder to see parallax in. If you do steps 1 through 4 you see fast images need to be shot so they can show their parallax.

And these 4 rules are the crude explaination of the rules Mr.Orman made. As described by me. If I'm wrong then please Mr.Orman, correct me.

I read Mr.Orman say the viewing angle should be 40 degrees, so I googled a calculator so the distance to see at that angle was easy to find; Link, for my 32", 16:9 aspect ratio TV, viewed at 3.1 feet, I get around 40 degree viewing angle.

So to put this calculator in context, you put in the info and get a 40 degree angle so you know how far you should be from your tv when you judge the parallax you want your content to have.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #21 of 56 Old 12-10-2010, 04:50 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post
What the rules mean, in a short crude description. Is the distance you see the screen, you need a certain amount of parallax.
This is so you see the picture in 3D.

If you have parallax but not enough for the distance the picture is seen, then you don't see 3D and it makes you feel odd.

So when you make the video content, you make the parallax fit a TV size, so
1.) you actually have the TV that is that size,
2.) you make the content with different parallax,
3.) you choose the parallax you feel comfortable with for that TV size viewed at that distance.
4.) you create content with that parallax.
5.) fast moving content is harder to see parallax in. If you do steps 1 through 4 you see fast images need to be shot so they can show their parallax.

And these 4 rules are the crude explaination of the rules Mr.Orman made. As described by me. If I'm wrong then please Mr.Orman, correct me.

I read Mr.Orman say the viewing angle should be 40 degrees, so I googled a calculator so the distance to see at that angle was easy to find; Link, for my 32", 16:9 aspect ratio TV, viewed at 3.1 feet, I get around 40 degree viewing angle.

So to put this calculator in context, you put in the info and get a 40 degree angle so you know how far you should be from your tv when you judge the parallax you want your content to have.
Sorry,
you must not edit anything if you are projecting on the screen that matches the original stereo window size.

You can float the stereo window at projection time if you are projecting on a screen that is different in size than the original target screen.

Mathew Orman
icester is offline  
post #22 of 56 Old 12-10-2010, 09:25 AM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by icester View Post

Sorry,
you must not edit anything if you are projecting on the screen that matches the original stereo window size.

You can float the stereo window at projection time if you are projecting on a screen that is different in size than the original target screen.

Mathew Orman

What I meant was there is no stereo window, and they the creator of the content, are deciding what the stereo window will be - what parallax the stereo window will be.

To decide the parallax of the stereo window, they first must own a TV the size they want to create the content for.

Then they need to sit at the distance from the TV equal to 40 degrees distance.

Then make the content with different parallax. Choose the parallax they want for their content. Then make their content with that stereo window/parallax.

This way, when they make their scene, if it is a fast scene. They can watch it on the TV at the correct distance and see if the stereo content if fusable comfortably. And if not then they adjust the scene creation so the person can comfortably fuse the scene.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #23 of 56 Old 12-10-2010, 09:38 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

What I meant was there is no stereo window, and they the creator of the content, are deciding what the stereo window will be - what parallax the stereo window will be.

To decide the parallax of the stereo window, they first must own a TV the size they want to create the content for.

Then they need to sit at the distance from the TV equal to 40 degrees distance.

Then make the content with different parallax. Choose the parallax they want for their content. Then make their content with that stereo window/parallax.

This way, when they make their scene, if it is a fast scene. They can watch it on the TV at the correct distance and see if the stereo content if fusable comfortably. And if not then they adjust the scene creation so the person can comfortably fuse the scene.

That is totally wrong and unacceptable.
Such techniques are use by gimmick content makers.

Mathew Orman
icester is offline  
post #24 of 56 Old 12-10-2010, 09:42 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
If you want to make content that is projection screen size independent then you must shoot with 65 mm IO parallel rig with at least 4k wide and 50 deg HFOV.
Then in post you can crop all to set desired stereo window or screen size.

Mathew Orman
icester is offline  
post #25 of 56 Old 12-10-2010, 09:54 AM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by icester View Post

That is totally wrong and unacceptable.
Such techniques are use by gimmick content makers.

Mathew Orman

Is it wrong because the parallax the content uses in this scenario is decided by the content maker and not the true stereoscopic parallax as you used in your horse riding picture?


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #26 of 56 Old 12-10-2010, 10:00 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

Is it wrong because the parallax the content uses in this scenario is decided by the content maker and not the true stereoscopic parallax as you used in your horse riding picture?

Parallax is created by scene content and if you edit it you will change the scale of the scene from real to gimmick.

So for example 1 inch cube sticking out the screen
will measure something different than 1 inch, when you stick a rule and measure it.

Realistic content means that scenes have the same dimension as in real world.

Mathew Orman
icester is offline  
post #27 of 56 Old 12-10-2010, 10:01 AM
AVS Special Member
 
Frank's Avatar
 
Join Date: Apr 1999
Location: Dept. of Offense
Posts: 5,194
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 14
Boy, I'm glad I can safely ignore all this when making my own 3D videos.

Did IQ's suddenly drop sharply while I was away?
I enjoy 3D in spite of HDMI 1.4!
Full screen only 3D doesn't cut it!
Frank is offline  
post #28 of 56 Old 12-10-2010, 10:08 AM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by icester View Post

Parallax is created by scene content and if you edit it you will change the scale of the scene from real to gimmick.

So for example 1 inch cube sticking out the screen
will measure something different than 1 inch, when you stick a rule and measure it.

Realistic content means that scenes have the same dimension as in real world.

Mathew Orman

I think your saying that realistic content has the parallax of realistic content. And that is something I chose to ignore as a requirement, but I see you don't see it that way.

IMO the parallax is allowed to be gimmick, or to say that differently, the content is allowed to be different than the real world. Because, the comfort of the viewer is the first rule. If the content shows the same when watching it on the TV as it looks in the real world, but the parallax is too large, for some like Frank it is uncomfortable. So, I decided to use the parallax of the stereo window to what the content creator decided it should be.
If you argue this post then please address the underlined part.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
post #29 of 56 Old 12-10-2010, 10:22 AM
 
icester's Avatar
 
Join Date: Apr 2010
Posts: 1,175
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
Quote:
Originally Posted by 8:13 View Post

I think your saying that realistic content has the parallax of realistic content. And that is something I chose to ignore as a requirement, but I see you don't see it that way.

IMO the parallax is allowed to be gimmick, or to say that differently, the content is allowed to be different than the real world. Because, the comfort of the viewer is the first rule. If the content shows the same when watching it on the TV as it looks in the real world, but the parallax is too large, for some like Frank it is uncomfortable. So, I decided to use the parallax of the stereo window to what the content creator decided it should be.
If you argue this post then please address the underlined part.

If you want comfort view the 2D version without any glasses.

There is nothing uncomfortable in realistic stereoscopic projection.

Mathew Orman
icester is offline  
post #30 of 56 Old 12-10-2010, 10:27 AM - Thread Starter
AVS Special Member
 
8:13's Avatar
 
Join Date: Sep 2006
Posts: 1,250
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 11
Quote:
Originally Posted by icester View Post

If you want comfort view the 2D version without any glasses.

There is nothing uncomfortable in realistic stereoscopic projection.

Mathew Orman

Fair enough. I will ask one more question and I need a technical answer.
We are talking about this rule from the first post:
"The view angle of the camera is the same as the viewer view angle."
Or as you said that the viewer see's the content with realistic geometry.

How would the film maker decide the stereo window parallax to use, as I think this description I quoted is very vague and nondescript.


There is new, and then you are new.
This is a moral of the bears and their cereal.
8:13 is offline  
Reply 3D Tech Talk

User Tag List

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off