Date   

Re: M4/3 cameras that will recognize the Panasonic 3D lens?

 

I see that the Panasonic document in the link is dated 8/2020,  so is it true that the GH5 and other more recent models DO have compatibility with and recognize the 3d lens? -Linda

On Sat, Sep 19, 2020 at 12:35 AM Linda N via groups.io <ljnygren=gmail.com@groups.io> wrote:
Thanks, Jeff.  I had looked at that page earlier but missed the h-f012 lens down near the bottom of the page. And the footnote  

The short list of excluded Panasonic cameras suggests that the others (that were released after the introduction of the 3D lens) SHOULD work . At least at the time this the list was compiled. But maybe not all later models such as the GH5??? Linda

On Sat, Sep 19, 2020 at 12:19 AM Jeff Ewen <jeffewen888@...> wrote:
https://av.jpn.support.panasonic.com/support/global/cs/dsc/connect/MicroFourThirds_lens.html

List of compatible Olympus cameras at bottom of page. plus the following reference to incompatible Panasonic cameras

It can not be used for the following Micro Four Thirds camera body (DMC-G1, DMC-G10, DMC-GF1,DMC-GH1). And all of Four Thirds body also.

Hope this is of some help

Jeff


Re: M4/3 cameras that will recognize the Panasonic 3D lens?

 

Thanks, Jeff.  I had looked at that page earlier but missed the h-f012 lens down near the bottom of the page. And the footnote  

The short list of excluded Panasonic cameras suggests that the others (that were released after the introduction of the 3D lens) SHOULD work . At least at the time this the list was compiled. But maybe not all later models such as the GH5??? Linda

On Sat, Sep 19, 2020 at 12:19 AM Jeff Ewen <jeffewen888@...> wrote:
https://av.jpn.support.panasonic.com/support/global/cs/dsc/connect/MicroFourThirds_lens.html

List of compatible Olympus cameras at bottom of page. plus the following reference to incompatible Panasonic cameras

It can not be used for the following Micro Four Thirds camera body (DMC-G1, DMC-G10, DMC-GF1,DMC-GH1). And all of Four Thirds body also.

Hope this is of some help

Jeff


Re: M4/3 cameras that will recognize the Panasonic 3D lens?

Jeff Ewen
 

https://av.jpn.support.panasonic.com/support/global/cs/dsc/connect/MicroFourThirds_lens.html

List of compatible Olympus cameras at bottom of page. plus the following reference to incompatible Panasonic cameras

It can not be used for the following Micro Four Thirds camera body (DMC-G1, DMC-G10, DMC-GF1,DMC-GH1). And all of Four Thirds body also.

Hope this is of some help

Jeff


Re: Bierstadt patented in cover book stereo viewer

robert mcafee
 

The White Mountains book used the Artotype process. My earlier posting gives some details of this process and a link where you can see some enlargements. 

On Friday, September 18, 2020, 07:18:26 PM EDT, Bob Aldridge <bob@...> wrote:


Interestingly, "Gems of American Scenery - White Mountains" was also published in 1878, and that - of course - was printed with a version of the "stochastic" technique (with a very fine random dot screen - almost like the grain of a lithographic limestone block). I have no idea whether this used "plates" as referred to by Joki, but clearly there was a viable printing technique available... It's actually not as crisp as the best of the albumen prints in the Frith book, but many of them are rather faded, so it's rather swings and roundabouts regarding the printed images in White Mountains vs the albumen prints in Frith. Of course, if you can unearth the original negative (as Denis Pellorin & Brian May have done for the Piazzi Smyth Tenerife book published in 1858) you can create much better prints :-)

Bob Aldridge

On 18/09/2020 21:53, robert mcafee via groups.io wrote:
Joki notes in 1878 no photo-mechanical half-tone process had been invented yet that could transfer original photos to plates for printing.  Therefore the reason to used tipped in photos.

Virus-free. www.avg.com


M4/3 cameras that will recognize the Panasonic 3D lens?

 

I have been unable to find a list of  M4/3 cameras that will "recognize" the Panasonic 3D lens. Do you have one, Francois? The GH2 and GX1,7,8 are the only ones I know of for sure.  And you mention the GH3 and GH4. Possibly the G2, GF2, and G3 , and the Olympus PEN EP-5 and the Olympus OM-D EM-1, based on info in reviews.  I thought I saw somewhere a reference to the Olympus OM-D EM-5 working with it, but I am less certain of that and cannot find the reference now. At this point I am wondering which cameras recognize the lens at all to use "auto mpo mode", even if the GH2 is the only model that gives mpos with 1920 wide images. 

The 3D lens should "fit" on any of the M4/3 cameras, and it will likely work OK in "taped contact mode" on all of them. If they are able to shoot  "without a lens", anyway. On my gx7, I was prompted to go into settings and enable "shoot w/o lens", and then the camera would capture the uncropped image pair (in RAW or jpg as selected), or can shoot video.  Linda

On Fri, Sep 18, 2020 at 9:21 PM Depthcam via groups.io <depthcam=yahoo.ca@groups.io> wrote:
On Fri, Sep 18, 2020 at 04:06 PM, Linda N wrote:
I am curious what the "automatic mpos" would show for pixel dimensions in 4:3 and also in 1:1 aspect ratio with your GH5

The problem there is that the 3D (MPO) mode (where the camera body recognizes the 3D lens) no longer exists on the GH5.  The GH3 and GH4 do recognize the 3D lens.  However they loose the multi-aspect ratio sensor so end up producing a file that is the same size as with the GX7.

Francois


Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

Depthcam
 

On Fri, Sep 18, 2020 at 04:06 PM, Linda N wrote:
I am curious what the "automatic mpos" would show for pixel dimensions in 4:3 and also in 1:1 aspect ratio with your GH5

The problem there is that the 3D (MPO) mode (where the camera body recognizes the 3D lens) no longer exists on the GH5.  The GH3 and GH4 do recognize the 3D lens.  However they loose the multi-aspect ratio sensor so end up producing a file that is the same size as with the GX7.

Francois


Re: Bierstadt patented in cover book stereo viewer

Bob Aldridge
 

Interestingly, "Gems of American Scenery - White Mountains" was also published in 1878, and that - of course - was printed with a version of the "stochastic" technique (with a very fine random dot screen - almost like the grain of a lithographic limestone block). I have no idea whether this used "plates" as referred to by Joki, but clearly there was a viable printing technique available... It's actually not as crisp as the best of the albumen prints in the Frith book, but many of them are rather faded, so it's rather swings and roundabouts regarding the printed images in White Mountains vs the albumen prints in Frith. Of course, if you can unearth the original negative (as Denis Pellorin & Brian May have done for the Piazzi Smyth Tenerife book published in 1858) you can create much better prints :-)

Bob Aldridge

On 18/09/2020 21:53, robert mcafee via groups.io wrote:
Joki notes in 1878 no photo-mechanical half-tone process had been invented yet that could transfer original photos to plates for printing.  Therefore the reason to used tipped in photos.

Virus-free. www.avg.com


Re: Bierstadt patented in cover book stereo viewer

robert mcafee
 

Correction Harroun & Bierstadt  patent (not Harround.)

Note Edward Bierstadt invented this improved stereoviewer integral to the book and Albert Bierstadt the painter was his brother.  Both made stereo images

American portrait and landscape photographer, and engraver. He was a brother of the painter Albert Bierstadt, and made a number of engravings of his brother's work. He opened his own studio in New York City in 1860. He also held an 1876 patent for an improvement to the Stereoscope Viewer, used for the extensive body of stereographs created by the Bierstadt brothers.

He was hired by William West Durant to take a series of photos for an advertising brochure entitled The Adirondacks, Artotype Views Among the Mountains and Lakes of the North Woods to publicize Blue Mountain Lake and Raquette Lake in the Adirondacks.

Much of his work is simply signed "E. Bierstadt NYC."


Some information on the Artotype "The Artotype is a form of collotype, a photomechanical printing process from a gelatin surface, that was available at Bierstadt's Artotype Atelier in New York.

Info below from site MetroPostcard Guide to Postcard Printing Techniques 2 .  Some enlarged images to be seen at this link (for Art Clark's question)

COLLOTYPE
The collotype is a continuous tone printing process first patented in France by Alphonse Louis Poitevin in 1855 under the name Phototypie. It began to be used commercially as the Albertotype in 1868 after Josef Albert in Germany perfected the method, but when patented in the United States one year later it was given the name Artotype. The technique begins with a greyed glass plate coated with a photosensitive dichromate colloid gelatin that puckers and cracks as it dries. When exposed to light through a reverse negative, the lit areas harden into an insoluble nonabsorbent finish. It is the areas within the reticulated cracks that harden the most because they are the thinnest part of the emulsion. They in turn will print the darkest in proportion to the tones of the original image. The dichromate emulsion in areas with little or no exposure to light remains soluble and is washed out from the gelatin with cold water. The plate is then printed in a similar manner to a lithograph. A solution of glycerin and water is spread over the plate's surface, which is absorbed by the remaining gelatin. Areas that are to carry the dark tones absorb little or no moisture while areas for the lighter tones and non-image areas absorb the most. When greasy ink is rolled over the gelatin on the plate, the non-image areas holding the most moisture repel the ink, and the dry hardened image areas attract the ink. Once printed the reticulated pattern creates a continuous toned image of incredible detail for which it is prized.

During the 19th century the collotype process was best known as Artotype, the name given to it by Josef Albert. Many others however elaborated on the process and trade names such as Hoeschotype, Inkphot, Paynetype, Photopane, Photo-type, and Albertype abounded. Some became more widely used than others and a number of different variations on this method were used to produce postcards. In 1868 the German Max Gemoser developed an important variation, which became known as Lichtdruck. The gelatin skin was removed from his processed plates and remounted onto litho-stones for greater support during printing. Despite certain shared characteristics, collotypes are not always easy to identify do to natural variations in the drying gelatin and numerous trade secrets. With no set standard the most uncommon collotype patterns may be difficult to distinguish from photogravure.
Collotype: The fine grain of collotypes with their white centers have a barnacle like look since the ink prints from the reticulated surface the plate. Though similar to the grain of aquatint and variable in size, the wormy look caused by the gelatin curdles is usually a dead giveaway in identifying the technique. This however can only be observed under extreme magnification as the close and closer details below of the postcard above demonstrate.
The greyed glass plate used to make collotypes was a major drawback for the process; it is very fragile and its shallow gelatin surface can rarely yield more than 500 impressions. Some have claimed to have pulled 2000 images from a single collotype plate but this may be an exaggeration. For large orders a new substrate would often have to be prepared substantially adding to the cost. Gelatin plates did not age well due to their organic nature and could not be stored for long periods to print at a later date. They were routinely scraped down after printing so that the glass could be reused. This severely limited the commercial applications of this process, but it proved adequate for small press runs and was widely used in postcard production. Glass was used as a substrate because it was one of the few materials that gelatin would adhere to in a thin even coat, but after August Albert discovered a way to mount a collotype emulsion onto an aluminum plate in 1896, this medium became easier to use and was finally adapted to the rotary press. Glass plates were still preferred by some printers who continued to use them into the 1960’s. Even when working with metal plates, the delicate gelatin emulsion still needed to be treated more carefully than other substrates, which slowed down production time. Even factors such as high humidity could severely limit their production; and they were usually printed in places where the weather was more stable. Despite all these drawbacks, collotype was still used to produce more black & white postcards than any other technique.
There were no special presses built to print collotypes, perhaps due to all the secrecy that surrounded this process. The image is usually transferred to paper on a modified lithography press or sometimes on a flatbed cylinder press. Its use in commercial printing greatly expanded after 1873 when a way was found to print collotypes with steam powered presses. Though it still remains the most accurate reproductive printing method available today, the process was largely abandoned in the 1990’s in favor of high resolution digital technologies. An order placed for real photo postcards in the early 20th century might be filled with printed collotypes nearly half the time because the resulting images were so similar that the two methods were generally considered interchangeable.



Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

Oktay
 

On Fri, Sep 18, 2020 at 03:30 PM, John Hart wrote:
"Why would you want such a "multi aspect sensor"?  I would prefer to use full width of a 4:3 sensor"

Because when you lose in height, you gain in width if you have a multi aspect sensor camera.
When you shoot a scene where you need to include  more of the width , you may shift to a wider proportion if you are ready to sacrifice the height.

3:2 has a little bit more width than 4:3,
16:9 has a little bit more width than 3:2.

The image area is indirectly proportional though.

It's 22.5mm diagonal of rectangles that is common in all aspect ratios.

Oktay


Re: Bierstadt patented in cover book stereo viewer

Bill Burns
 

Here's an 1880 pamphlet from Harroun & Bierstadt advertising the Artotype process:

"Artotypes are photographs in printing-ink"

https://digital.clarkart.edu/digital/collection/p1325coll1/id/2102

Artotype appears to have been similar to the 1855 collotype printing process:

https://en.wikipedia.org/wiki/Collotype

I have not yet found an 1876 patent for Bierstadt.

--
Bill

On 18-Sep-20 4:53 PM, robert mcafee via groups.io wrote:
David,
Unfortunately the auction house shows only the exterior binding.  In the paperback reprint of this book, edited by Robert Joki, there are a couple pictures of the inside cover with integral viewer (poor copy attached).  That inner cover is stamped (in gold I presume) "Patented March 21st 1876 Manufactured by Harround & Bierstadt Proprietors N.Y."
As for number of copies, the Onondaga Historical Society has a copy, Syracuse University has a copy, the editor Robert Joki has a copy and the auction house has a copy - so at least 4 survive (assume none of this were Mr. Joki's copy).  Mr. Joki estimated the selling cost at that time as likely around $100 (quite a sum for 1878).  He suggests it was probably sold by subscription and it was common to charge a wealthy family or business to feature an image of their home/business and to sell them a copy of the book.
H. Perry Smith produced the text.  Photographers were Myron Judd and William McLeish.  Printing and binding was done by Syracuse printer Truair, Smith, & Bruce.
Joki notes in 1878 no photo-mechanical half-tone process had been invented yet that could transfer original photos to plates for printing.  Therefore the reason to used tipped in photos.
The auctioneer's site has a couple scans of stereo pairs from the book. Joki un-bound his own copy to scan the images (as some pages had separated due to the tight binding).  One reason I would like to locate Mr. Joki is to find out if he has scans of the stereo pairs he might share or sell.
So ... This is definitely the Bierstadt Patent design.
Bob McAfee
Syracuse, N
Inline imageInline image

--
Bill


Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

 

Hi, John. I would want it just for the 3D lens, since the GH2 gives  slightly higher resolution images from the "automatic mpo" function compared to standard sensor models, and the theory is that it might do the same in other models if they had this type of sensor. Even the 20 mp sensor of the gx8 returns only 1824 x 1368 per side mpo images (for unknown reasons), while the multi aspect sensor on the 16+ mp GH2 results in 1920 x 1440 images. Somehow the GH2 seems to be unique in this regard, presumably due to the multi-aspect sensor. Some of us are "lazy" and prefer to get usable images right out of the camera. But 1920 wide would be nice rather than 1824. 

It sounds as if you routinely use your  3D lens with contacts covered or otherwise inactivated so that you can use video? And you also get stills of  2000+ on each side with the 20mp GH5, after editing? Have you tried the camera for stills with an unmodified (contacts) 3D lens? Perhaps it will capture a wider pixel resolution image even though the 20 mp gx8 does not do any better than the gx1 & gx7. I am curious what the "automatic mpos" would show for pixel dimensions in 4:3 and also in 1:1 aspect ratio with your GH5. Linda



On Fri, Sep 18, 2020 at 3:30 PM John Hart <hart3d@...> wrote:
Why would you want such a "multi aspect sensor"?  I would prefer to use full width of a 4:3 sensor for all aspect ratios higher (like 16:9, 3:2) => better resolution.
This is what GH5 does:  4:3 is 5184x3888, 16:9 is 5184x2920, 3:2 is 5184x3456 but 1:1 is 3888x3888 (as constrained by height of sensor).
For video the full sensor is used but the output is resized to the mode (HD, UHD, C4K, 6K).

With just 3D-lens on washers, and no toe in prism effect, I typically get almost HD (~1850wide per side) video for my subjects AFTER registration using C4K mode.  Stills are easily above 2000 wide for most subjects not too close.  Real close requires toe in to get above HD.  If you had perfect toe in, the LR resolution would be about 2400 x1350 or 2400 x 1800 for 16:9 (my standard use) and 4:3, after cropping out the septum effect.

HTH,  John                        


Re: Bierstadt patented in cover book stereo viewer

robert mcafee
 

David,
Unfortunately the auction house shows only the exterior binding.  In the paperback reprint of this book, edited by Robert Joki, there are a couple pictures of the inside cover with integral viewer (poor copy attached).  That inner cover is stamped (in gold I presume) "Patented March 21st 1876 Manufactured by Harround & Bierstadt Proprietors N.Y."

As for number of copies, the Onondaga Historical Society has a copy, Syracuse University has a copy, the editor Robert Joki has a copy and the auction house has a copy - so at least 4 survive (assume none of this were Mr. Joki's copy).  Mr. Joki estimated the selling cost at that time as likely around $100 (quite a sum for 1878).  He suggests it was probably sold by subscription and it was common to charge a wealthy family or business to feature an image of their home/business and to sell them a copy of the book.  

H. Perry Smith produced the text.  Photographers were Myron Judd and William McLeish.  Printing and binding was done by Syracuse printer Truair, Smith, & Bruce.  

Joki notes in 1878 no photo-mechanical half-tone process had been invented yet that could transfer original photos to plates for printing.  Therefore the reason to used tipped in photos.

The auctioneer's site has a couple scans of stereo pairs from the book.  Joki un-bound his own copy to scan the images (as some pages had separated due to the tight binding).  One reason I would like to locate Mr. Joki is to find out if he has scans of the stereo pairs he might share or sell.

So ... This is definitely the Bierstadt Patent design. 

Bob McAfee
Syracuse, N

Inline imageInline image



Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

John Hart
 

Why would you want such a "multi aspect sensor"?  I would prefer to use full width of a 4:3 sensor for all aspect ratios higher (like 16:9, 3:2) => better resolution.
This is what GH5 does:  4:3 is 5184x3888, 16:9 is 5184x2920, 3:2 is 5184x3456 but 1:1 is 3888x3888 (as constrained by height of sensor).
For video the full sensor is used but the output is resized to the mode (HD, UHD, C4K, 6K).

With just 3D-lens on washers, and no toe in prism effect, I typically get almost HD (~1850wide per side) video for my subjects AFTER registration using C4K mode.  Stills are easily above 2000 wide for most subjects not too close.  Real close requires toe in to get above HD.  If you had perfect toe in, the LR resolution would be about 2400 x1350 or 2400 x 1800 for 16:9 (my standard use) and 4:3, after cropping out the septum effect.

HTH,  John                        


Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

Ronald Schalekamp <info@...>
 

Thanks Linda,

I tried both. I had the Loreo Macro 3D lens on a Canon DSLR. Made a focusing aid myself and used a ring light; i was okay with the results.
Later i changed from Canon to Panasonic mirrorless so the Pana 3D lens was a great thing for me. 
Using it with washers. Have to try adding a closeup lens also.

I feel the results are somewhat the same up to now as with the Loreo lens. Both need post-processing, but i've gotten quite handy with that.

Enclosed some images i shot with the Loreo 3D lens, quite some time ago.
And the last 2 images of the Dragonflies were shot with the Pana 3d lens.

I've never tried a macro box; it just doesnt appeal to me. Setting up both cams, quite a large box, keeping the mirrors clean.
The ease of use is what speaks for the Pana lens.

It is a pity you cant get a shallow depth of field using beamsplitters.
Again, doing the chacha is the best choice i feel, when nothing is moving. We easily forget about that.

And sorry for my earlier trouble replying to messages. Im a bit new at this.

Greets, Ron


ve a sh



Another formerly popular 3D system discontinued. Why?

John Rupkalvis
 


Re: Automatically colourise B&W images using AI

John Rupkalvis
 

Often with both colorizing and 2D to 3D conversions the best procedure is to use the automatic software as a "starting point" and then go from there.  An imperfect partial result can save time, and then following this with manually "finishing" can yield superior results.  

John A. Rupkalvis
stereoscope3d@...

Picture


On Fri, Sep 18, 2020 at 9:45 AM Depthcam via groups.io <depthcam=yahoo.ca@groups.io> wrote:
On Fri, Sep 18, 2020 at 10:41 AM, David Starkman wrote:
However, considering what these programs can do in less than a minute, sometimes (not always) the results are amazingly good.

Hi David,

As I wrote in my post, I tested quite a number of pictures with these AI online colorizing sites and indeed, some results were excellent.  However, I often found that some of the colorizing was partial even when the rest of the picture looked pretty good.  So there was still a need to do some manual adjustments.  In fact one of the clips in my links shows how to use an AI colorized image as a starting point on which to build.  He even shows how to bring an image back to high resolution by extracting the color from the conversion and reapplying it to the original b&w image.

But it still remains that I tried quite a number of pictures using these programs and my best way to describe the results is "hit & miss".

Francois


Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

 

Here is a diagram that I think explains how the multi aspect sensor works (also attached):

Unfortunately I think the GH2 is the only Panasonic micro four thirds camera with 16 mp (or greater) that has this feature. The GH1 and GH5s do have the feature (as do some of their non-interchangeable lens cameras), but fewer megapixels (12 for the GH1, 10 for the GH5s) . If anyone knows of other models in the Panasonic or Olympus lines that might have this feature, and so might give the 1920 x 1440 images from the 3D lens, let us know. Linda


On Fri, Sep 18, 2020 at 12:01 PM Depthcam via groups.io <depthcam=yahoo.ca@groups.io> wrote:
On Fri, Sep 18, 2020 at 03:54 AM, Laurent DOLDI (Toulouse, France) wrote:
Maybe the GH2 upsizes these 3D images to fit full HD width?

I doubt it.  I think the answer lies in the GH2's sensor being described as an 18mp multi-ratio sensor.  Even though the output is 16mp, it adjusts the aspect ratio over an 18mp sensor.

Here is an extract from DPreview:

"In the years since Olympus and Panasonic jointly announced the Micro Four Thirds standard there have been eight models launched to date, seven of which share the same 12MP sensor. The only exception was the GH1 which, despite its 12MP output actually featured a 14MP sensor. The sensor was slightly larger than the format's imaging circle, which allowed it to deliver images in different aspect ratios while retaining the same image diagonal. And now the GH2 follows the same pattern, shooting 16MP images using its brand-new, similarly over-sized 18MP sensor."

Francois
_._,_._,


Re: Bierstadt patented in cover book stereo viewer

Bob Aldridge
 

Don't forget the Frances Frith book "Egypt, Nubia and Ethiopia" that was published in 1862!

No viewer (well, one was offered as a separate item: "The Stereoscopes specially adapted for use with this volume, compactly folded in an elegant pocket-book form, fit for the drawing room table, price 5s. and upwards, may be had of Messrs Negretti and Zambra)

It was illustrated with 100 albumen prints glued down to the pages! Quite a bit is known about the process and even who made the prints - but I've forgotten! :-) However, as well as a slightly incomplete copy of the book (with colour photocopies in place of the missing pages), I also have 30 or 40 of the print pairs that never made it into books...

Anyway, this domonstrates that creating books with a lot of photographically illustrated images was perfectly feasible.

And, judging by the price of the viewer (5 shillings in 1962 is worth about £30 or $40 today...) the book was probably reasonably priced - but I don't know what the print run was. 100, perhaps? or 1000? I, personally, know of a few copies, so not as rare as the Syracuse book.

Bob Aldridge

On 18/09/2020 17:16, David Starkman via groups.io wrote:
As for the Syracuse book currently listed at $7,500, the seller indicates that possibly only 7 copies are known to have been made. Looking at the photos, and the fact that it indicates that 103 mounted albumen print stereoviews were in the book, this looks more like a book that was practically hand-made, using the maker of one of those fancy Victorian photo albums to make the basic book, and all of those prints having to be hand pasted in. I'm attaching a copy of an 1871 patent for a Stereoscopic Photo album that had a built-in viewer, and could have been the basis for this "book" design. We have one of these 1871 stereoscopic albums, and is only one of two that we have ever seen. 
 
As you have said, the Syacuse book must have been quite costly to produce, even in 1878!  Too bad that the reprint book did not choose to print the full stereo views!!!

Virus-free. www.avg.com


Re: Panasonic Lumix 3d lens with Close-up Lenses vs Spacers Test

Depthcam
 

On Fri, Sep 18, 2020 at 03:54 AM, Laurent DOLDI (Toulouse, France) wrote:
Maybe the GH2 upsizes these 3D images to fit full HD width?

I doubt it.  I think the answer lies in the GH2's sensor being described as an 18mp multi-ratio sensor.  Even though the output is 16mp, it adjusts the aspect ratio over an 18mp sensor.

Here is an extract from DPreview:

"In the years since Olympus and Panasonic jointly announced the Micro Four Thirds standard there have been eight models launched to date, seven of which share the same 12MP sensor. The only exception was the GH1 which, despite its 12MP output actually featured a 14MP sensor. The sensor was slightly larger than the format's imaging circle, which allowed it to deliver images in different aspect ratios while retaining the same image diagonal. And now the GH2 follows the same pattern, shooting 16MP images using its brand-new, similarly over-sized 18MP sensor."

Francois


Re: Automatically colourise B&W images using AI

Depthcam
 

On Fri, Sep 18, 2020 at 10:41 AM, David Starkman wrote:
However, considering what these programs can do in less than a minute, sometimes (not always) the results are amazingly good.

Hi David,

As I wrote in my post, I tested quite a number of pictures with these AI online colorizing sites and indeed, some results were excellent.  However, I often found that some of the colorizing was partial even when the rest of the picture looked pretty good.  So there was still a need to do some manual adjustments.  In fact one of the clips in my links shows how to use an AI colorized image as a starting point on which to build.  He even shows how to bring an image back to high resolution by extracting the color from the conversion and reapplying it to the original b&w image.

But it still remains that I tried quite a number of pictures using these programs and my best way to describe the results is "hit & miss".

Francois

2361 - 2380 of 129371