Is R3D 4:4:4 or 4:2:2?

[T. Singh]

New research on random color filter arrays

New research on random color filter arrays

Last few years have seen some serious research on random RGB Color Filter Arrays (CFA). In 2009, a French scientist, Laurent Condat, proposed a random RGB CFA in L. Condat, “A NEW RANDOM COLOR FILTER ARRAY WITH GOOD SPECTRAL PROPERTIES,” IEEE Transactions on Image Processing, 2009. This CFA was designed to optimize performance of the demosaickers available then.

While this was a big step forward the result was not competitive with Bayer. Subsequently, in 2011, a co-author and I presented a improved demosaicker in Singh, T and Singh, M, "Disregarding Spectral Overlap - A Unified approach for Demosaicking, Compressive Sensing and Color Filter Array Design," in Proc. of IEEE ICIP (2011). When combined with Laurent Condat's Random RGB CFA this algorithm outperforms Bayer systems demosaicked with the best of the published algorithms. As expected, the random RGB system shows remarkable resistance to aliasing. It does so by randomizing the aliased signal, making it look and behave like noise.

For more information and test images see this page.

Disclosure: This work is in the last stages of research now, and we plan to commercialize it in the near future.

 

[Eric Lange]

Last few years have seen some serious research on random RGB Color Filter Arrays (CFA). In 2009, a French scientist, Laurent Condat, proposed a random RGB CFA in L. Condat, “A NEW RANDOM COLOR FILTER ARRAY WITH GOOD SPECTRAL PROPERTIES,” IEEE Transactions on Image Processing, 2009. This CFA was designed to optimize performance of the demosaickers available then. While this was a big step forward the result was not competitive with Bayer.

Subsequently, in 2011, a co-author and I presented a improved demosaicker in Singh, T and Singh, M, "Disregarding Spectral Overlap - A Unified approach for Demosaicking, Compressive Sensing and Color Filter Array Design," in Proc. of IEEE ICIP (2011). When combined with Laurent Condat's Random RGB CFA this algorithm outperforms Bayer systems demosaicked with the best of the published algorithms. As expected, the random RGB system shows remarkable resistance to aliasing. It does so by randomizing the aliased signal, making it look and behave like noise.

For more information and test images see this page.

Disclosure: This work is in the last stages of research now, and we plan to commercialize it in the near future.

Anyone from the Medium Format market shown an interest in your de-mosaicker?

Thinking especially from the point of view of 80MP sensors and chromatic aberration etc.

Having a cursory look at your web site and info… what exactly is meant by “quality can be traded off with computation”? (in your case).

Also wondering for technical applications for example in photogrammtery are the spatial positions of features and targets more spatially accurate for fine features that would normally be susceptible to aliasing artifacts. For example in stereo–photogrammetric applications it’s good to have very decent spatial fidelity between left and right images for very fine (high frequency) corresponding elements and features.

For some of our applications we use high res lossless compressed imagery, it sounds like the algorithms that you have might produce smoother features at the finest levels of detail with acceptable contrast. For very fine stereoplotting work and stereo machine algorithms this might provide an easier input image for very high res apps that mainly use stills or low frame rate applications.


Mr Singh, sounds very interesting,

Cheers,

Eric


p.s. This caught my eye... when you said “Disclosure: This work is in the last stages of research now, and we plan to commercialize it in the near future”… I only wish our R&D phases and “commercialization” phases were so discrete… to us R&D never stops and I think it’s obviously the same at RED… (he, he) although we don’t get to get to all the things we would really like to… but one foot in front of the other and keep going.

 

[Wayne Morellini]

This US patent Application filed in 2007 (from Japan) describes a very particular set of solutions to problems normally associated with diagonally arranged photo-sites on a CMOS imaging array.
See

http://www.google.com/patents?id=xV-iAAAAEBAJ&printsec=abstract&zoom=4#v=onepage&q&f=false

You can click on the full citation and I urge folks to look through all of the diagrams at very least.

This might answer a few questions and also put a few things to bed… until physical testing is possible.

From Summary of the Invention:

“As described above, lines in a CMOS image sensor are arranged in the vertical and horizontal directions with respect to the chip or image pickup area. However, when miniaturization of pixels is promoted, or when a pixel transistor is shared by a plurality of photo diodes, light incident on each photodiode may be obstructed, particularly in a CMOS image sensor having a diagonal pixel array. In addition, vertical and horizontal lines (for example, the vertical lines 16 and 17 in FIG 23) limit wiring layout possibilities, and thus the area occupied by photodiodes has to be reduced.

Con’t “[0013] The present invention has been made in view of the above circumstances. Accordingly, there is a need for a solid-sate pickup device with a diagonal pixel array which prevents obstruction of light incident on a photoelectric conversion unit and thus allows an increase in a light-recieveing aperture ratio of photoelectric conversion unit.…

…
[0015]… “With this arrangement, obstruction of light collection of photoelectric conversion unit can be prevented.”

I.e. for diagonally arranged photo detectors.

This document also cites other variants of diagonally arranged arrays, but this one claims to solve more of the problems normally associated with diagonal arrays.

Clearly, the way these things are actually physically manifested as very complex layered architectures and shapes of photo sites and wiring are about a million miles away from the near infantile marketing materials and diagrams that one normally finds.

What I’m saying here , even if you were to three dimensionally re-construct everything presented in the above patent application in a very high end multi-physics simulation software there is still no way that one would be able to actually extrapolate the actual performance of the system described. Really!

I would gently point out perhaps that maybe we should desist from extrapolating very complex solid state quantum behaviors, DSP and performance from lousy marketing materials.
It’s terrific that everybody here is super smart and very interested and I too have been guilty when the F65 was first announced to also sketch out on the back of an envelope the expected characteristics of the F65… but in reality this is a bit of a folly, but never the less is good to get some sort of appreciation of how these things might “tick”.


In this sense I feel for Graeme somewhat that he has to very patiently field such queries and dispel potential FUD from other marketing claims. That’s a lot of hours taken out of his day and I guess such marketing from other quarters can become a little irritating after a while.

I am very grateful to Graeme over the years to kindly educate me on this and that and gain a greater appreciation of almost a more “holographic” understanding of the interaction between various arrays and sampling strategies as being a much more complex and subtle macroscopic process. He has changed my perception a great deal not to think of photosites in such a “discrete way”. Thanks for that!

Maybe I foresee a “thread closed” in the not too distant future just so Greame can get on with his day job! (he, he).

So I eagerly await an input image and output image set of tests also for the F65… it’s awesome there is community that can pull together to perform such test. Overall its great (I think) that the whole 4k ecosystem is gathering momentum and investment for the long haul…. Every player is important to help make that happen including RED’s (supposed competitors).

Still I don’t know what 4:4:4: means anymore? … and is it a useful idea?

Cheers,

Eric

Foot note, I'm not saying this is how exactly the F65 functions but is used as an example to illustrate the true complexity of such a sensor and that it is not possible to extrapolate performance from marketing diagrams or even the best information available of the true architecture of such a system.

Well, I hope they are "super smart" enough to realise that I have only been addressing potential angular demosacing advantages that we can examine and do something about, rather than stupid circuite abstracts we can't even know for certain at the moment

4:4:4 usefullness? Not so smart, it is an useful practical mathematical boundary. Enough coffee Eric, plenty of sleep. The reason binary works is that it is as deterministic as 4:4:4 done right boundaries (think beyond 3 dimensions).

 

[Wayne Morellini]

Thinking out loud...

Potential direction in imaging and display technologies.

Somebody intelligent. Thanks again Hrvoje, nice. I had thought up many patterns years ago, something like this too I think. One was concentric boxes of different colors to do 4:4:4 (of different thickness if nobody has instantly figured that out yet). This is not ideal (colour response instead of filter, response to small passing detail reduced but not eliminated, shrinking boxes result in higher response). I think that I came up with repeating cycle of concentric lines to solve some issues. But circles, X, + patterns with repeating fills would work. The repeat lines are connected together as one well, but obvious problems without sufficient performance. On the "surface" of it, that was a singe direction. Of course a series of verticle or horizontal lines has issues, but a cross pattern doesn't (an array of primaries per pixel), but how to make, and efficently. Going back to my very first hyposis for a sensor in the 1990's, there might be away without filters. Despite his accurate 4:4:4 can be over Bayer, really dumb people will not realise you don't need all that extra performance, or Bayer, to produce a nice cinema image. But they are caught in their quality fantasies. But, 4:4:4 full range, full boundary, tells you what is really there more than Bayer, so is useful. Enough clues.

On these patterns, I can see making photosite subsegments as potentially useful if you want this ability. You reconfigure and relink pad segments to make new unuts according to need, with full enough fill. But of course how do you do this dynamically interframe


Of course, somebody tell people the solution to circuites and the incidence of light, currently moving from consumer to pro.

 

[Wayne Morellini]

Good night, I'll let go of the sides of the mouse maze now.

 

[Eric Lange]

Well, I hope they are "super smart" enough to realise that I have only been addressing potential angular demosacing advantages that we can examine and do something about, rather than stupid circuite abstracts we can't even know for certain at the moment 4:4:4 usefullness? Not so smart, it is an useful practical mathematical boundary. Enough coffee Eric, plenty of sleep. The reason binary works is that it is as deterministic as 4:4:4 done right boundaries (think beyond 3 dimensions).

These types architectures have already been built…

I think what’s interesting, that kinda adds “weight” to the Bayer layout is that it dosen’t matter too much how your photosites are laid out. It seems you can’t cheat physics… (at least not yet…). I have to admit on an intuitive level I like the random layout idea… its almost film like in terms of grain like concept, but as I said before I would imagine that a large redundancy of photosites vs effective final output res would be required. In the MF world having 60MP to 80MP plus sensors really does challenge the best lenses out there and hence software that is able to effect a really good guess at how to regroup apparently “fringed” features seems to have a genuine utility.

Here’s one for Graeme… what’s difference between having an ordered Bayer array in the first place where you then partially randomize the signal/capture data in a localized way and then process, versus having a randomized array that gets processed anyway. Aren’t we just adding noise to break up higher contrast artifacts then smoothing and then re-grouping (in most of these examples)?
The other thing is that a lot of what is being discussed does not take into account compression methods of any kind especially for dynamic imagery.

It is an art… there’s no about that.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Wayne you should publish a book of all of your ideas a sort of Davinci Codex of our time…

I know it would be millions of diagrams and notes etc…

I think that would be cool as an object in its own right… and a peak into the mind of Wayne Morinelli…

What’s your drawing like…?

Cheers,

Eric

 

[Hrvoje Simic]

“As described above, lines in a CMOS image sensor are arranged in the vertical and horizontal directions with respect to the chip or image pickup area. However, when miniaturization of pixels is promoted, or when a pixel transistor is shared by a plurality of photo diodes, light incident on each photodiode may be obstructed, particularly in a CMOS image sensor having a diagonal pixel array. In addition, vertical and horizontal lines (for example, the vertical lines 16 and 17 in FIG 23) limit wiring layout possibilities, and thus the area occupied by photodiodes has to be reduced.

Con’t “[0013] The present invention has been made in view of the above circumstances. Accordingly, there is a need for a solid-sate pickup device with a diagonal pixel array which prevents obstruction of light incident on a photoelectric conversion unit and thus allows an increase in a light-recieveing aperture ratio of photoelectric conversion unit.…

Hexagonal arrangement I proposed has R, G & B photodiodes in horizontal sets, with each color occupying a separate row.
While allowing an increased aperture ratio.

RGBW version is a "bit" more complex.
With a couple of bonuses.

 

[T. Singh]

Anyone from the Medium Format market shown an interest in your de-mosaicker?

Thinking especially from the point of view of 80MP sensors and chromatic aberration etc.

Having a cursory look at your web site and info… what exactly is meant by “quality can be traded off with computation”? (in your case).

Also wondering for technical applications for example in photogrammtery are the spatial positions of features and targets more spatially accurate for fine features that would normally be susceptible to aliasing artifacts. For example in stereo–photogrammetric applications it’s good to have very decent spatial fidelity between left and right images for very fine (high frequency) corresponding elements and features.

For some of our applications we use high res lossless compressed imagery, it sounds like the algorithms that you have might produce smoother features at the finest levels of detail with acceptable contrast. For very fine stereoplotting work and stereo machine algorithms this might provide an easier input image for very high res apps that mainly use stills or low frame rate applications.


Mr Singh, sounds very interesting,

Cheers,

Eric


p.s. This caught my eye... when you said “Disclosure: This work is in the last stages of research now, and we plan to commercialize it in the near future”… I only wish our R&D phases and “commercialization” phases were so discrete… to us R&D never stops and I think it’s obviously the same at RED… (he, he) although we don’t get to get to all the things we would really like to… but one foot in front of the other and keep going.

Medium Format cameras are indeed a very good fit for Chromatic Aberration correction, as is any system where the lens is at a distance from the sensor. DSLRs, for instance, need such lenses to make room for the pentaprism/mirror. These lenses suffer from more chromatic aberration. How have Canon mount users' experience been?

Best,
Tripurari

 

[Wayne Morellini]

These types architectures have already been built…

I think what’s interesting, that kinda adds “weight” to the Bayer layout is that it dosen’t matter too much how your photosites are laid out. It seems you can’t cheat physics… (at least not yet…). I have to admit on an intuitive level I like the random layout idea… its almost film like in terms of grain like concept, but as I said before I would imagine that a large redundancy of photosites vs effective final output res would be required. In the MF world having 60MP to 80MP plus sensors really does challenge the best lenses out there and hence software that is able to effect a really good guess at how to regroup apparently “fringed” features seems to have a genuine utility.

Here’s one for Graeme… what’s difference between having an ordered Bayer array in the first place where you then partially randomize the signal/capture data in a localized way and then process, versus having a randomized array that gets processed anyway. Aren’t we just adding noise to break up higher contrast artifacts then smoothing and then re-grouping (in most of these examples)?
The other thing is that a lot of what is being discussed does not take into account compression methods of any kind especially for dynamic imagery.

It is an art… there’s no about that.

>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>>

Wayne you should publish a book of all of your ideas a sort of Davinci Codex of our time…

I know it would be millions of diagrams and notes etc…

I think that would be cool as an object in its own right… and a peak into the mind of Wayne Morinelli…

What’s your drawing like…?

Cheers,

Eric

Film, as we think of it in past cinema, is not authentic. For others I'm not explaining, I know Eric knows why explicitedly. Whatever technique is used, unless it maps to fullrange reality, or our reality (fooling) of 4:4:4 full range, it is not autentic. It does not matter how close it is to human vision, that is not the question, it has to contain enough information to fool human vision, however we are looking at it in a reasonable environment, therefore tge ability to emulate 4:4:4 full range, that Bayer presents some ability to do. So, much argument here over the week by various people, has been non authentic

Maybe you jest Eric, at least get the name right. I have the volume of new ideas to start making one, and can spontaneously form a new authentic idea every second or less, when working and well, and refuse to dwell there otherwise, more authentic volume than anybody I have ever heard of. It is not the art of producing spontaneouse rubbish, but of producing well filtered, well worked out, practical authentic reality. Every simple thing may include a multtude of aspects they answer (simply), otherwise they may not be simple in practice for users.

 

[Eric Lange]

Film, as we think of it in past cinema, is not authentic. For others I'm not explaining, I know Eric knows why explicitedly. Whatever technique is used, unless it maps to fullrange reality, or our reality (fooling) of 4:4:4 full range, it is not autentic. It does not matter how close it is to human vision, that is not the question, it has to contain enough information to fool human vision, however we are looking at it in a reasonable environment, therefore tge ability to emulate 4:4:4 full range, that Bayer presents some ability to do. So, much argument here over the week by various people, has been non authentic

Maybe you jest Eric, at least get the name right. I have the volume of new ideas to start making one, and can spontaneously form a new authentic idea every second or less, when working and well, and refuse to dwell there otherwise, more authentic volume than anybody I have ever heard of. It is not the art of producing spontaneouse rubbish, but of producing well filtered, well worked out, practical authentic reality. Every simple thing may include a multtude of aspects they answer (simply), otherwise they may not be simple in practice for users.

Sure but Wayne I have worked with spectrophotometers and similar systems that have a full curves for almost all wavelengths practically for each “pixel”… for conservation applications. When you look at these full spectral curves you can only ask “what does it all mean”? There is really no display device on the planet that can really make such data truly “perceivable”?

So for example when I worked at the Getty and at The Smithsonian, for example on a single oil painting we would do X-rays, Auto radiographs, IR short and long wavelength as well as spectrophotometer and other types of color measurement and UV imaging as well, to basically pin down certain chemical analysis or compositions or reveal under drawings and so on. Every imaging system has to have a “focused” purpose.

For commercial applications in film and television a system that has a really good dynamic range that fits the useful gamut of human perception and current and future display technology and enough dynamic range that is useful for color grading in the DI suite is perhaps good enough? I personally don’t see the need for absolute (full spectral) colorimetry for shooting movies … Do you?

With kind regards,

Eric

p.s. sorry I spelt your name wrong...

 

[Wayne Morellini]

Sure but Wayne I have worked with spectrophotometers and similar systems that have a full curves for almost all wavelengths practically for each “pixel”… for conservation applications. When you look at these full spectral curves you can only ask “what does it all mean”? There is really no display device on the planet that can really make such data truly “perceivable”?

So for example when I worked at the Getty and at The Smithsonian, for example on a single oil painting we would do X-rays, Auto radiographs, IR short and long wavelength as well as spectrophotometer and other types of color measurement and UV imaging as well, to basically pin down certain chemical analysis or compositions or reveal under drawings and so on. Every imaging system has to have a “focused” purpose.

For commercial applications in film and television a system that has a really good dynamic range that fits the useful gamut of human perception and current and future display technology and enough dynamic range that is useful for color grading in the DI suite is perhaps good enough? I personally don’t see the need for absolute (full spectral) colorimetry for shooting movies … Do you?

With kind regards,

Eric

Oversampling to get authentic reality. It is good you are questioning, from that you can decide you can go down to achieve something, or up to get something more, a lot of people will lock in to the complete fate of the method (faith, like being on Red user a bit) or defocus (loose capacity or mental illness). Me, I'm focused on results, what's there and what's further. So, we have two contrasting personalities, and abilities. Good, we could probably compliment each other. My wondering/scanning attention can nail onto possibilities in the processes/data, yours see that there is more and pick things up. A ultimate secret, one for the history books, is that in ways, in a physical sense, everything is process. That opens up all the treasures of engineering. Still, an interesting aspect just popped up in my mind, does that mean that randomness is spiritual, in the theory that spiritual comes from randomness, unengineerable in itself, and spontaneously forming acts. Some ideas that I have looked at (out of assessment rather than belief in).

I knew that it was spectral chemical analysis. To decipher the authentic history and technique, and good conservation of an item. I could get into more but have some work I want to patent in the area. Display devices don't matter so much, they can emulate through narrow band frequencies, but we need to work with the broad band to decide what to do with the narrow band.

I get you, no I'm not advicating an absolute need for full spectral recording, just acknowleding the reality that it is more authentic than full range 4:4:4 recording that I do advocate as being more authentic than others, and we try to emulate with Bayer. Dwindling returns. If I did not, the analysis would be too gappy, and somebody could object that full spectral would be even more authentic. Covering the bases. I know, a real pain (notice I didn't say 'all bases' as there maybe stuff I missed, or I should say 'something/s'. Nice being able to write spontaneously ). However, if you could see everything I'm working on, you would be stunned.

 

[Eric Lange]

Oversampling to get authentic reality. It is good you are questioning, from that you can decide you can go down to achieve something, or up to get something more, a lot of people will lock in to the complete fate of the method (faith, like being on Red user a bit) or defocus (loose capacity or mental illness). Me, I'm focused on results, what's there and what's further. So, we have two contrasting personalities, and abilities. Good, we could probably compliment each other. My wondering/scanning attention can nail onto possibilities in the processes/data, yours see that there is more and pick things up. A ultimate secret, one for the history books, is that in ways, in a physical sense, everything is process. That opens up all the treasures of engineering. Still, an interesting aspect just popped up in my mind, does that mean that randomness is spiritual, in the theory that spiritual comes from randomness, unengineerable in itself, and spontaneously forming acts. Some ideas that I have looked at (out of assessment rather than belief in).

I knew that it was spectral chemical analysis. To decipher the authentic history and technique, and good conservation of an item. I could get into more but have some work I want to patent in the area. Display devices don't matter so much, they can emulate through narrow band frequencies, but we need to work with the broad band to decide what to do with the narrow band.

I get you, no I'm not advicating an absolute need for full spectral recording, just acknowleding the reality that it is more authentic than full range 4:4:4 recording that I do advocate as being more authentic than others, and we try to emulate with Bayer. Dwindling returns. If I did not, the analysis would be too gappy, and somebody could object that full spectral would be even more authentic. Covering the bases. I know, a real pain (notice I didn't say 'all bases' as there maybe stuff I missed, or I should say 'something/s'. Nice being able to write spontaneously ). However, if you could see everything I'm working on, you would be stunned.

Wayne you should absolutely do that.

The Australian patent office is pretty awesome. The examiners that we had really knew their onions and asked really excellent questions and granted one of our submissions/applications in Oz.
The thing that’s really great about the Australian Patent office much more so than the majority of other patent offices in different territories is that they really go the extra mile to help individual Australian inventors. In other words they have excellent and accessible resources to really help the “home grown” Australian (individual) inventor. Also a lot of the materials and forms they provide is written in very plain easy to understand English… no legalese or government entity “speak”.

Clearly you (Wayne) have an extremely fertile mind. From personal experience I have learnt that yup in my life time I have had many thousands of seemingly “great” ideas, probably a few hundred or less are truly viable and probably a short list of about twenty are really worth it and then you have to face the reality that you are only going to be able to pursue only one or two really fundamentally new ideas. Then with all your heart, will and whatever resources you can muster, you put all your “wood” behind one major idea and maybe two more minor ones with the understanding that the others are probably going to be on the back burner for a very long time.

Legally having an idea does not mean that you own it… even having a patent on something does not even automatically grant you the right to “practice the invention” in some cases.

Surely out of the tens of thousands or even hundreds of thousands of viable ideas that you have had why not file an application on one or two with the Australian patent office? Their filling fees are very reasonable and their deadlines and schedules etc are VERY flexible indeed. At least if you for example you don’t follow through on all of your applications it would still constitute prior art against larger entities that might choose to seek a patent on a similar invention.

All your wood behind one idea that might be good for Australia and beyond… why not?

The Australian patent office is VERY keen to promote Australian inventions/inventors.

Yours,

Eric

 

[Rob Ruffo]

I admit my question was a bit cheeky, because I knew it was an altogether different question, but I also knew it would get the sort of responses I was looking for, which is that:

A) Theoretically modern RGB compression could do a better job than 4:2:2 downsampling pre compression, but hell, 4:2:2 is fast, known and visually pretty hard to spot so heck, so rather than bemoan the world we live in going with that for now makes sense.

and

B) Bruce might write up his 4:2:2 Open EXR workflow in a post, which is a workflow that makes a heck of a lot of sense in my head theoretically, but isn't one that would be considered the 'norm' in tradition heavy iron type workflows where a lot of the literature comes from (especially not at 4K resolution.) and one that I would need to do a bit of research to implement. Bruce, if you are ever in Auckland, New Zealand, let me buy you a beer.

Having only really had my head in the post game for about 5 years, and having spent much of that scrounging around for scraps of information on prosumer centric forums or boffiny personal websites of video engineers to get to grips with a lot of these concepts (because somewhat surprisingly at first I discovered even a lot of people who worked day to day in industry really had no idea about what was actually going on under the hood and would spout misinformation at the drop of a hat.) I am truly thankful for the quality of communication and openness with information that is on Reduser!

It's not that hard to spot 422 when you have greenscreen, or even an aggressive secondary color grade based on HSL qualification.

 

[Wayne Morellini]

Wayne you should absolutely do that.

The Australian patent office is pretty awesome. The examiners that we had really knew their onions and asked really excellent questions and granted one of our submissions/applications in Oz.

They are pretty good, but increasingly polluted by Us and world patent standards. I used to be a member of the Queensland inventors association of Australia, when it was being run by a professional and pretty good. Since then, the legal disclosure paper work has not been kept up to with the law (as required under law and how the law prescribes). I have not bothered to go back, you can guess why. While there is a noble tradition of inventing in Australia, it has had a non professional, no legally correct, hobby aspect to it, and rip off merchant engineers.

I am in no way been fit to see through a patent on my own, to have a New York law firm try to bust it because I am not a company with lots of money to further legal action. The recommendation was to work out all the business plan and marketing, as well as technical, seek an interested substantial player that can afford to do it, get them under agreement to disclose, show it and try to reach a deal. I think also filling before hand. As well, unlike America, we are not allowed to write an application ourselves, but have to hire a patent attourney specifically to do it, from memory. A lot of hard work, time and money, oodles of time. Some things the US gets right, but others are just expliotive against the rights of true indivduals in favor of companies. Nor have I been fit enough to fight an action against two groups it appears have directly used and patented a couple of my world leading technologies, which became or are regarded as such. I have had a turn this afternoon, from finally doing well to feeling like throwing up sitting here writing this before. The process of improving my physical health is taking time.

So, I have sought partnerships to give me the breathing space, or just take over. But every single person I have had to try to work with lands up being an untrustworthy total let down, I do the work, they do precious little to nothing, and are so degenerate they can't do anything towards a certain reward many times their earning capacity. Now the obstacles are so high, a lot of the easy/cheap opportunities are gone, and the money required so much, just to be able to partnership to hand it over to an large established player in the market is a challenge. If I had the knowledge, making space vehicles would probably be easier than some technological things. Not to say I am not working towards something, but getting real.

I should have to do no more than say something publically, or in private in relation to anybody that hears/sees etc either officially, accidentally or covertly, to establish a prior art date (generally, or in relation to the parties that have seen/head etc it in private) and the ability to file on those disclosures within a period (if not indefinite). There should be a method to lodge ideas before filling for protection, allowing it to attract business contracts to pay for the process. But the law is as it is.

The thing that’s really great about the Australian Patent office much more so than the majority of other patent offices in different territories is that they really go the extra mile to help individual Australian inventors. In other words they have excellent and accessible resources to really help the “home grown” Australian (individual) inventor. Also a lot of the materials and forms they provide is written in very plain easy to understand English… no legalese or government entity “speak”.

Probably in no small part due to my old mob.

Clearly you (Wayne) have an extremely fertile mind. From personal experience I have learnt that yup in my life time I have had many thousands of seemingly “great” ideas, probably a few hundred or less are truly viable and probably a short list of about twenty are really worth it and then you have to face the reality that you are only going to be able to pursue only one or two really fundamentally new ideas. Then with all your heart, will and whatever resources you can muster, you put all your “wood” behind one major idea and maybe two more minor ones with the understanding that the others are probably going to be on the back burner for a very long time.

Eric, you know how much it costs to do a patent spread year after year. An Australian patent is not worth much on it's own, I might spend more money trying to stop grey imports from China than the patent could generate. You have to file across the main markets within something like a year or so, which includes manufacturing nations like China, just to make it more tight. Not, to mention all the dubious patents you might cross that are putting their hands out because you violated something as non original as a business practice patent on putting your hand out would be (thanks America).

Most of my ideas are viable and worth it, the subsequent pickup and success rate by others is proof (some sensor pattern ideas I mention are not so worked out, but abstracts looking at how detail can be picked up, as I was not making sensors). My mind works in a pretty practical reductionist way homing in on target. Imagine having 6 months of work and insight/revelation going through your head over four hours writing non stop just trying to keep up, all ordered, all worked out, all issues problems and enhancements worked out for the whole area, not just one product. That is the other issue, I can go through everything in a sitting, and I don't have one patent in front of me but even dozens. I know the ideal way to make it and the moment I file a base patent on that ideal way, other patents that stem from it will become obvious to people to file. A nightmare. The reason why, are the solutions are often eventually simple in the end, once you work through them, appropriate simplification often increases efficency, but the juxtapositions are where the complex solution work is done to make the solution simple. This is something often missed in the bloated dumb systems we have these days, a sheer lack of eligence.

Legally having an idea does not mean that you own it… even having a patent on something does not even automatically grant you the right to “practice the invention” in some cases.

I know, but until somebody else discovers it, it is a trade secret. It is always mine in a fashion, exclusive patent rights aside, but not exclusively after disclosure.

Surely out of the tens of thousands or even hundreds of thousands of viable ideas that you have had why not file an application on one or two with the Australian patent office? Their filling fees are very reasonable and their deadlines and schedules etc are VERY flexible indeed. At least if you for example you don’t follow through on all of your applications it would still constitute prior art against larger entities that might choose to seek a patent on a similar invention.

Over 15 years ago people had spent $250,000 over here to develop and patent individual products. The patent is only the start and then you have your patents for other markets and periodical fees, as mentioned above. If I have such trouble with health, how am I going to push through complex paper work and legal processes, which is one of the most complex things to do for me. I can write newer better processes, maybe a hundred times easier than doing the process, those are two sections of the brain, ones got it tough. Not even my family, that could easily afford to block the offending patent application to bring the other party to the table, could be bothered even though the rewards were more than they have ever earnt (what the patent holders are doing is only scrapping the financial potential, but they can't see). As the runt of the litter, I have suffered most aggrieviously. Sure I'm pessimistic, now!

All your wood behind one idea that might be good for Australia and beyond… why not?

Yes, I know, that is what I thought too, but even all the business people I know are so dumb as not to get that. I'll tell you what, by the looks of it, I walked past Bill Gates in the street and didn't recognise him due to overhead lighting exaggerating his glasses and face, though I knew he was familiar, as I did not find out he was in town till the next day. It now turns out he was here due to a major third world disease research project being conducted around my local town. It is a dream to be able to run into Bill Gates and talk to him about my revolutionary third world computer/free communications platform project, for economic expansion, safety and social benefit. I had been contemplating wherever to take it to Google, as they are geared to take advantage of the new technology, as a sort of mini alternative to the android ecosystem, but Bill's charity is also a good fit. Just typical how things go.

I wanted to be able to afford to make a company to exploit the inventions, to provide finance for works towards the public good, virtually solely.

If you think it is such a great idea Eric, you are welcome aboard, I'm sure you can afford it.

Yours,

Eric

PS: Been side tracking patent issues with the new third world OS, relying on copyright protection techniques, (but still some patents unfortunately). As the old OS would had meant dozens of patents, from new ways around normal os issues, scheduling, instruction set design, networking, graphics, programmable hardware language, just about everything I could lay my mind upon.

The new OS and platform promises some revolutionary new advances in computing technology. There is more than one way to skin a cat.

 

[Wayne Morellini]

Now, I forgot to reply to a color space issue mentioned before. I have noticed massive inconsistencies in dealing with color space from one device to another. What we need is a Mega color space container, that covers all we can see, or more, and a mapping from the recording device into that color space to translate the recorded values to this space. This will make footage more portable, so that footage is recorded with original values and translated during processing, or for display, to new mappings Footage will be smaller than translating to the Mega color space in the first place, as that will probably require 20 bit+ or real numbers.

 

[Graeme Nattress]

It's not that hard to spot 422 when you have greenscreen, or even an aggressive secondary color grade based on HSL qualification.

Back in 2001, Pananvision with Sony at NAB did a demo of just that - 4:4:4, 4:2:2, 3:1:1 (HDCAM) keying. I'd just finished keying some 3:1:1 for the other demo they were showing - computer graphics depth of field matched to lens metadata.

The idea was that most people couldn't tell which key was which. And indeed "most people" couldn't. I got all three correctly named. You've just got to know what to look for.

The situation isn't as bad today though because keying software has improved and it either incorporates, or people know to use, some kind of chroma resolution fixing filter.

Graeme

 

[Wayne Morellini]

http://www.eoshd.com/content/498/meet-the-prosumer-camcorder-which-is-4k-for-2k
http://www.eoshd.com/content/6795/4k-for-4k-and-why-you-probably-shouldnt-get-a-jvc-hmq10
http://www.eoshd.com/content/6824/j...cinema-world-with-interchangeable-lens-camera
http://www.eoshd.com/content/6983/nikon-may-add-4k-video-to-future-cameras-j1-sensor-already-capable
http://www.eoshd.com/content/6976/canon-developing-4k-raw-video-format
http://www.eoshd.com/content/6817/s...lls-4k-video-off-sensor-in-400-compact-camera

Tick, tock. Pity about the low end.

 

[Gunleik Groven]

Now, I forgot to reply to a color space issue mentioned before. I have noticed massive inconsistencies in dealing with color space from one device to another. What we need is a Mega color space container, that covers all we can see, or more, and a mapping from the recording device into that color space to translate the recorded values to this space. This will make footage more portable, so that footage is recorded with original values and translated during processing, or for display, to new mappings Footage will be smaller than translating to the Mega color space in the first place, as that will probably require 20 bit+ or real numbers.

LOL

ACES?

 

[Wayne Morellini]

www.eoshd.com/content/7120/canon-5d-mark-ii-mosaic-engineering-vaf-5d2-anti-aliasing-filter-review

Canon video filter solution.

 

[Wayne Morellini]

LOL

ACES?

Looked it up.

At least they are catching up

But, I thought the range that was being used on film was more limited than what I was suggesting (gone full visible spectrum by the looks of it, good, can I have more please), as a general reel to television/consumer monitor format, rather than just cinema? This was a suggestion that could maintain the original files data density and integrity, by keeping it the same until changed (or rerendered in it's color space until it is exported to final product). Does aces allow that?

Looks like one less thing to do