What do you guys make of this?
Claudio
"Amarison" <amar...@hotmail.com> schrieb im Newsbeitrag
news:UqmdnQuYL9i...@rogers.com...
And this PDF date 21/06/2005 Achieves key milestones in PIXe
Commercialisation:
http://www.asx.com.au/asxpdf/20050621/pdf/3r8jhs8tdspr6.pdf
Their website is down since Nov 23, 2004 (typical what happens when you
compression to much :-))
http://web.archive.org/web/*/http://www.blazelimited.com
Part text The Age member link:
Blaze chief executive Peter Jon Hartshorne made the claim yesterday and
said patents protecting the intellectual property in the US and
elsewhere were now assured.
Part text The Age member link:
Pixe is a pre-compression filter that, used in conjunction with a
standard compression codec (compression/decompression) such as MPEG2 or
MPEG4, allows a video file to be reduced to one-450th of its original
size.
Pixe reduces a file by up to nine times before it is compressed in the
usual way, usually to a maximum of 50:1 by one or other of the 30-odd
standard codecs in use around the world. The file is decompressed after
transmission by reversing the process.
In this way, Mr Hartshorne said, real-time video files could be made
small enough to suit the limited computational power of mobile handsets
or transmitted to computers or settop boxes over low-rate
telecommunications lines.
Pixe could deliver broadcast quality video in real time over standard
DSL (copper telephone) lines, not possible with current technologies.
Part text The Age member link:
Pixe's underlying technology was developed in 2003 by Peter Bevan, a
Perth reprographics operator. "Our process changes the way pixels are
arrayed on the screen, producing a smaller picture before (standard
compression software) is applied."
Blaze evaluated Pixe in August 2004 and provided seed capital to prove
the concept before fully acquiring it.
"This is not a replacement technology. It does not change the existing
standard or proprietary codec. That is very important.
"If we interfered with (standard compression processes) Pixe would not
be adopted by the larger companies in the world," Mr Hartshorne said.
Who's next?
yeah, maybe someone could copy/paste the article into a post or something...
This new link maybe:
http://www.heraldsun.news.com.au/common/story_page/0,5478,15689838%255E664,00.html
Regards,
Wimjan
--
Sleet is van vroeger, tegenwoordig gaat alles kapot.
and blazelimited.com site seems to be down. When searching google for
blazelimited.com, google does not show the Cache link, could it be that the
website has been down for a long long time?
"cr88192" <cr8...@NOSPAM.hotmail.com> wrote in message
news:97354$42b88db2$ca80102a$24...@saipan.com...
"Wimjan" <ne...@inventio.nl> wrote in message
news:MPG.1d22b8563...@news.leidenuniv.nl...
Thanks now I found their patent number:
System and Method for Transforming Graphics Data
Patent number: AU2005201173
Publication date: 2005-04-07
Inventor: BEVAN PETER THOMAS
Applicant: PIXE PTY LTD
Classification:
- international: G06T3/20
- european:
Application number: AU20050201173 20050318
Priority number(s): AU20040901429 20040318; AU20050201173 20050318
Here the full patent application (97 pages, happy reading!):
http://pericles.ipaustralia.gov.au/aub/pdf/nps/2005/0407/2005201173A1/2005201173.pdf
Maybe not so bad deal if it does what they claim:
Blaze International Ltd. PIXe Pty. Ltd. Video compression & imaging
software $7,000,000
http://www.techdealmaker.com/041015/
I have tried to understand the document.
My first impression is, that the main idea of the
patent consists of:
1. dividing each video frame into red, green, blue
color components
2. choosing a segmentation of the reference frame
for all three color components independently of
each other
3. choosing a segmentation of the next frame
which can be different from the segmentation of
the reference frame (for all three color components
independently of each other).
4. defining of a mapping between segments of the reference
frame (for all three color components) and the segments
of the next frame
5. composing the next frame out of the data from the
reference frame using this mapping.
The color value of the pixel in the next frame is
composed of pixel values of another pixels in the
reference frame according to the mapping of the
segments of the reference frame to segments of
the next frame.
What I am missing here is how new data, which
can't be derived from the reference frame come
into the next frame and what is the advantage of
doing as described in the patent, so there is
probably a problem with my understanding of the
patent content :-( .
The usual problem with reading patents is, that from the
description of the claims it is still not clear, what is the core
idea behind it and what are the advantages. Only the claims
are defined, without further details revealing the true
background of the patent.
In this special case I can't see how the mapping should
be done to achieve high compression, so from what
I understand now, I see no chance even only to try to write
down an appropriate compression algorithm myself.
Any hints towards clarification?
Claudio
"Sportman" <spor...@gmail.com> schrieb im Newsbeitrag
news:1119399804.0...@f14g2000cwb.googlegroups.com...
I feel dubious on the whole thing.
Sportman schreef:
Sportsman, you're a fast-learner.
http://www.blazelimited.com/whatwedo-pixe.html
"PIXe exploits an innovative technique called pixel oscillation to create a
significant reduction in file size. Furthermore, as the PIXe technology is
an optical system, rather than a complex mathematical system, it can be
deployed into existing software compression methods to enhance and further
reduce file sizes."
--
Andrey Kuznetsov
http://uio.imagero.com Unified I/O for Java
http://reader.imagero.com Java image reader
http://jgui.imagero.com Java GUI components and utilities
Is it me, or are all of the latest wacko compression claims coming from
Australia (specifically Melbourne)? Adam's Platform, this, a few
others... Something's got to be in the drinking water down there :-)