Stereoscopic 3D at Home

1
Stereoscopic 3D at Home

• Sheila Frederixon, Matt Gillett, Amy Gracik
• Stereoscopics is the technology that combines two
  separate images to create a 3D image. It is the
  most used form of 3D from Cinemas, to home
  theaters, and video games This is possible due to
  the fact that human eyes collect light
  independently from each other, thus the image we
  see is the combination of two 2D images.

2
Using Shutters to view Stereoscopic images.

• Keys to Home Theater 3D
• Video must be captured with two cameras or
  separated in complex computer renderings
• The video is sent to the TV as two separate
  images that alternate at a rate of 240Hz.
• The Liquid Crystal Shutter glasses close and open
  independently at 120hz in order to separate the
  observers image.
• The glasses and TV alternate between left and
  right images together via an IR, or Bluetooth
  link
• Your brain and eye can not operate quick enough
  to separate the two images, so they are combined
  into a Binocular 3D image (allows you to see
  depth)

3
How Shutter Glasses Work

• Since the human eye operates between 48 and 60Hz
  (depending on study)
• Shutter glasses manufacturers design the glasses
  to operate at 60Hz.
• That is the lenses turn on and off in 1/60th of a
  second.  
• But when the lens is open for that 1/60th of a
  second the LCD lens is clear for
• ½ and shaded for the other.  There fore it acts
  more like 240Hz

This is why you need a 240Hz TV in order to get
a truly smooth 3D image. (See movie)
4
How to Develop a 3D image with Cameras
  These two equations represent the perspective
position of two 2D points with respect to the 3D
space point. A and A are 3x3 matrices that
specify the intrinsic parameters of the first
camera with respect to the second. Pn is a 3x4
matrix that normalizes the perspective
projection. Here D represents a 4x4 matrix
containing a rotation, R, and a translation, t.
This transforms the 3D point from the real world
coordinate system into a camera coordinate
system.   If we rearrange the first equation and
make it dependent on the depth value Z, you
get     If you then substitute the new equation
into the second equation, you can find the depth
dependent relation between the two perspective
views and the same 3D scene.  
5
When Creating a Stereoscopic Image
On a stereoscopic 3D TV display there are two
slightly different perspective views of a 3D
scene. This means there is a slightly altered
variant to the previous equations. There are
still two cameras being used, however the
difference lies in the part of the 3D scene that
is going to be reproduced exactly on the display
screen. This mathematically is done via
shift-sensor approach. This is formulated as a
displacement of the cameras principal point. This
means that there is a horizontal shift of h, in
the secondary camera   In this case also,
RI, where I is the identity matrix. This lets
simplifications be made to the original 3D
imaging equation we found to become
6
References

• Fehn, Christopher. "Depth-Image-Based Rendering
  (DIBR), Compression and Transmission for a New
  Approach on 3D-TV." Fraunhofer-Institut FÄur
  Nachrichtentechnik, Heinrich-Hertz-Institute.
  Print.
• Ozaktas, Haldun M, and Levent Onural. Three-dimens
  ional Televison Capture, Transmission, Display.
  Berlin Springer, 2008. Print
• http//www.youtube.com/watch?vB9kHiJ2kvDQfeature
  channel Why 120Hz is not enough