 |
| |
Canvid
Video Codec Whitepaper
| |
Modern
Techniques
Shannon
Theory: Modern Video Transmission and Video Storage
for television broadcast and movie storage are
based upon “Shannon theory” (Claude
Shannon 1948). This theory treats a video image
merely as binary digits (bits). It disregards
the contents of the image and determines the bit
rate solely on hardware parameters such as image
size, resolution and scanning rate. The images
shown on the screen are irrelevant such that a
random noise image requires the same bit rate
as a blank image. |
|
|
| |
| Cosine
Transform
This
video compression technique was conceived
by a French mathematician, and thus
it is named the Fourier Transforms.
Cosine transforms are used in JPEG
compression for still images, MPEG-2
compression for moving video. In this
technique compression process selectively
removes information that is deemed
to be least important to the human
eye. In order to attain high compression
rates, large amount of information
needs to be removed. This process
leads to lower quality image. Removing
information removes detail resolution
in the images and introduces image
artifacts. Compression achieved using
this techniques consistently produces
fuzzy objects and thus questionable
image quality. |
Wavelet
In
wavelet compression a single video
image is sub-divided into several
images and some of the sub-divided
images are dropped and the remaining
sub-images are recombined to produce
original image. The inherent problem
with this technique is that large
number of sub-images needs to be dropped
to attain high compression ratios.
The higher the drop, the lower is
image quality. Therefore, this technique
can not produce commercially useful
compression ratios without significant
distortion. |
|
|
|
|
| |
Fractal
Compression: The Fractal Compression
is a forging technique which generates images
that look approximately like the originals. The
output images are mere approximations of the input
images. Such forgeries can produce good quality
images some of the time but not all the time.
These shortcomings of the modern techniques had
been the biggest hurdle in delivering high quality
video images over the Internet. Broadcast of video
or streaming of video over the Internet requires
large compression and all current techniques provide
poor quality and fuzzy video images when subject
to large compression.
This has been the trap for video transmission
over the Internet and video storage. The modern
techniques , at high compression ratios, can not
provide high quality video images for transmission
and storage.
|
|
|
| |
 Canvid
Advantage
Canvid Video Compression Solution is different.
It treats the contents of the image rather than
the image itself.
Canvid is built on the theory
that, you only need to carry forward video data
from one picture frame to the next picture frame
which is not already known to the next picture
frame. Everything already known is redundant and
need not be constantly carried forward. The images
of a picture frame are compared with the images
of next frame to locate pixels that have changed
brightness. Every pixel that has not changed is
ignored. The changed pixel are addressed, treated
using cosine transform, wavelet or fractal compression.
This radically different approach leads to the
production of high quality video images at unprecedented
compression rates and can be easily transmitted
over the Internet.
|
|
|
| |
Setting
New Benchmarks
Canvid represents the biggest technological break
through in video compression, transmission and
storage. It provides distinct advantages over
the modern techniques by providing high compression
ratios, low bit rates, high quality video images
and security. It is the ultimate solutions to
diverse applications employing rich media for
video transmission and video storage.
|
|
|
| |
 Compression
The compression is defined as the ratio between
the size of digitized raw video file to the size
of its compressed video file. The higher the compression
ratio, easier it is to transmit, and store. A
video picture 352x288 pixels in size with three
color separation and 8 bit precision occupies
311 Kbytes. If sequence of these pictures is sent
uncompressed at 24 pictures per second it will
require Internet speed of 60,000 kilo bits per
second. A normal CD-ROM can store only about 2
minutes of such television. This requires enormous
compression ratios which cannot be achieved with
acceptable image quality using conventional compression
methods. Internet video is currently of very poor
quality with extremely jerky motion. Even the
HDTV standard approved by the Federal Communications
Commission (FCC) produces rather blurred images
with a jumpy flickering motion that is almost
dizzying.
To transmit standard television signal over 56
kbps dial up modem requires 1200:1 compression
ratio. Until the development of Canvid , it was
literally impossible as the existing techniques
allow compression ratios of 150-200:1 before the
image is highly distorted. Canvid has achieved
average compression ratios of 400:1. For certain
applications the compression ratios has been as
high as 1500:1. Canvid can store 3 full length
movies on standard CD-ROM and 2 full length movies
on 256MB flash card of standard PDA.
|
|
|
| |
Broadband
The broadband for video transmission is defined
as bit rate (bits per second). The standard bandwidths
available include Dial-up modem speeds of 28 kbps
and 56 kbps, Cable with 400 kbps, DSL and T1 with
1.5 Mbps and Local Area Connection (LAN) with
10/100/1000Mbps. Nearly 80% of world’s 2
billion Internet users access the Internet through
a PC local LAN or Modems.
Canvid
is the only solution that provides streaming video,
real time video broadcast and video conferencing
at or below Modem connection speeds.
In
Canvid the bit rate is dependent only on motion
within the images. Canvid further capitalize on
the fact that the human eye can either perceive
fine color resolution of still image or rapid
movement in the image but not both at the same
time.
Canvid
has consistently demonstrated the following unmatched
performance when it comes to transmission of video
images over the Internet.
• Streaming video: 10 kbps
• Video Conference: 28 kbps
• Video Broadcast: 56 kbps
|
|
|
| |
| |
Quality
The quality of a compressed image is largely
dependent on whether the compression technique
is “lossy” or “Lossless”.
The three most commonly applied techniques
include ‘Cosine Transform’, Wavelet
Image Compression’ and ‘Fractal
Image Compression’. All these three
techniques are “lossy Techniques”.
Canvid
delivers high quality images regardless
of the image resolution, bit rate and color
precision.
|
|
 |
|
|
|
|
| |
Security
Canvid has built-in encryption option which allows
secure video conferencing via the Internet or
satellite without the possibility of unauthorized
interception. It largely solves the security problems
associated with the Internet today.
|
|
|
| |
Summary
Canvid offers:
• High quality video images
• High degree of compression ratios
• Transmission of video over Modem connections
• Storage and display of video images on
standard CD-ROM, servers, Flash cards
• Streaming of video over internet
• Broadcast of television over the Internet
• Broadcast of live events over the Internet
• Real time video conferencing between two
parties over the Internet
• Real time video conferencing between two
parties or multiple parties over LAN
|
|
|
| |
|
|
| |
Contact
Information
Contact
CSC Global Technologies Inc. for further information.
Call
403.275.5804
|
|
|
|
| |
Specifications
Video
Output
AVI, ASF
Interleave: 1 second, ½ second, 1 frame
Tracks
Video and Audio
Image
Size:
40x30 to 1600x1200 pixels
Constraints: Aspect ratio or Multiple of 4
Other features: Blur, Unsharp mask, Adaptive noise
reduction, Watermarks
Adjustments
Gamma,
Brightness, Contrast, Black and white restore,
Hue, Saturation
Compression
Bit
Depth: 32, 24 or 16 bit color
Frame Rate: 7.5 to 30 frames per second
Keyframes: 1 to 10,000
Bit Rate: 1 to 10,000 kbps
Encoding: 1 – pass CBR, 1-pass quality,
1-pass quantizer, 2-pass – 1st pass, 2-pass-2nd
pass int. & ext.
Quantizer: Maximum I Frames 31, Minimum I Frame
1
Maximum P Frames 31, Minimum P Frames 1
Settings: Motion – None to very high
Quantizer – MPEG, MPEG – Custom, Modulated
Max. I Frames –10,00
Min. I Frame – 1
Overflow: Max. overflow – 100%
Min. overflow – 0%
Max. bit rate – 10,000
Compatibility
Tested
with Adobe Premiere 5.x and up
Tested with Terran Media Cleaner 4 and up
Tested with VirtualDub 1.4.7 and Up
Tested with Windows Media Encoder 4
Tested with Windows Media Encoder 8 (AVI) only
Tested with Windows Media Player 9 (AVI)
System
Requirements
Windows
32 Bit O/S (Windows 95/98/ME/NT/2000/XP)
64
Megabytes of System Ram (128 MB recommended for
compression)
Microsoft
DirectX Version 8 or above
Video
color depth minimum 256 (16-bit color recommended)
Pentium
166MHZ or greater CPU (233MMX minimum recommended)
|
|
|
| |
|
|
|
|
|
|
|
|
|
