Computational Photography

Started 2 months ago | Discussions
chipmaster
chipmaster Senior Member • Posts: 2,987
Computational Photography
1

An interesting article on the new phones., after all this site is about digital photography.

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

Now the laggards have for the most part beaten the incumbents for total % of pictures taken by the majority.

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

https://www.nytimes.com/2019/10/15/technology/personaltech/google-pixel-photography.html

-- hide signature --

" Today's Pictures Are Tomorrow's Memories "

 chipmaster's gear list:chipmaster's gear list
Nikon 1 V1 Nikon Df Nikon 1 V3 Nikon D810 Nikon D5 +21 more
pixseal
pixseal Veteran Member • Posts: 3,604
Re: Computational Photography
6

I'm the first with the popcorn!

-- hide signature --
 pixseal's gear list:pixseal's gear list
Canon EOS 7D Mark II Canon EOS Rebel SL2 Canon EF-S 10-22mm F3.5-4.5 USM Canon EF-S 17-55mm f/2.8 IS USM Canon EF 100mm F2.8L Macro IS USM +18 more
PJPfeiffer Contributing Member • Posts: 846
Re: Computational Photography

chipmaster wrote:

An interesting article on the new phones., after all this site is about digital photography.

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

Now the laggards have for the most part beaten the incumbents for total % of pictures taken by the majority.

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

https://www.nytimes.com/2019/10/15/technology/personaltech/google-pixel-photography.html

I wonder if a similar discussion took place about Kodak when digital entered photography?

 PJPfeiffer's gear list:PJPfeiffer's gear list
Panasonic Lumix DMC-FZ300 Snapseed +3 more
mosswings Veteran Member • Posts: 9,449
Re: Computational Photography
3

chipmaster wrote:

An interesting article on the new phones., after all this site is about digital photography.

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

Now the laggards have for the most part beaten the incumbents for total % of pictures taken by the majority.

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

https://www.nytimes.com/2019/10/15/technology/personaltech/google-pixel-photography.html

You have a rather conspiratorial perspective on why computational methods appeared first in smartphones and not traditional cameras.

Your first speculations were roughly in the ballpark - computational methods appeared in smartphones first because they had to do something to improve their image quality vis a vis large-sensor camera systems.  Then the manufacturers figured out what to do with all that computational power to add some cool features and convenience items.

Large sensor cameras have not had to use computational methods exactly because they're large sensor cameras and the intrinisic quality of image is much greater.
Now as for laggards beating the incumbents, these are totally different markets you're talking about. People have always just wanted to take pictures, and smartphones allowed them to do something that traditional cameras are not designed to do - easily share images as part of an ephemeral conversation stream. Smartphones are always with you and the images you can take with them can interact with everything else that the smartphone does to enhance your everyday life.

Traditional cameras are about making art and telling a particular kind of story, much like writing a book is.  Most people don't want to do that.

So far, it's Sony and the u4/3 crew that seems to be applying computational methods first.  But remember, the first thing they have to solve is to make the workflow easier. That's communication and configuration methodology, far less compute intensive than what Google and Apple are doing.

 mosswings's gear list:mosswings's gear list
Olympus XZ-1 Olympus Stylus 1 Nikon D90 Nikon D7100 Sony a6400 +7 more
WarrenPeas
WarrenPeas Contributing Member • Posts: 747
Pipe Dream think a week would go without this topic.......
2

chipmaster wrote:

An interesting article on the new phones., after all this site is about digital photography.

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

Now the laggards have for the most part beaten the incumbents for total % of pictures taken by the majority.

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

https://www.nytimes.com/2019/10/15/technology/personaltech/google-pixel-photography.html

-- hide signature --

IG: @DooDooYellow

 WarrenPeas's gear list:WarrenPeas's gear list
Sony RX100 VA Sony a7R III Sony a6400 Sony E 16-50mm F3.5-5.6 PZ OSS Sony E PZ 18-105mm F4 G OSS +6 more
j3ffw
j3ffw Regular Member • Posts: 125
Re: Computational Photography
2

chipmaster wrote:

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

or a lack of info (small sensor) and an output jpg that could do the processing in a timely manner

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

The article states hdr that wasn’t an phone camera exclusive  one could also say focus stacking, pixel shift, a stitched panoramic file.  So to jump ahead, me personally I want the raw file, what will be nice is more advanced apps that one can program to set shooting sequences, iows an app that can run shooting scripts through the camera.  You are right why have the manufacturers not added more sequences than just hdr and pixel shift.  When will hdr pixel shift and focus stacking work in tandem?  I’m sure the answer is once you code the software for it to work

For one thing are the software companies able to handle it?  Capture one can’t do pixel shift, not great at panorama stitching or even hdr combinations.  While you can’t really blame the camera manufacturer there is no doubt all of their apps are lacking

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

this is a funny statement because I have to ask, do you even know? What is zoom, what is low light action?

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

what has google done? I don’t want a heavily processed jpg that can only hold up to viewing on a 6” phone with a limited attention span.  Seems like the next major step will be when  Capture One or “that which shall not be named” add these features to their software.

-- hide signature --

My Architectural Photography Portfolio
http://www.jeffwolfram.com

Lee Jay Forum Pro • Posts: 55,002
Re: Computational Photography
1

The purpose of computational photography is to compensate for not having the optics needed to do the job. For that reason...

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ?

...hopefully no one.

My 2015 SLR has panorama, stacking for noise reduction and HDR, and I don't use any of it because I use the optics I need instead.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
Tuloom Veteran Member • Posts: 3,319
Re: Computational Photography
1

I'd rather some AI in my ecosystem.

But with visionaries like Lee Iacocca now gone, I don't think we have a chance.

As Lee once said to crowd of drive thru daiquiri employees, "No AI, No Care; I'll take a Strawberry Margarita."

Reilly Diefenbach
Reilly Diefenbach Forum Pro • Posts: 13,517
Re: Computational Photography

what has google done? I don’t want a heavily processed jpg that can only hold up to viewing on a 6” phone with a limited attention span. Seems like the next major step will be when Capture One or “that which shall not be named” add these features to their software.

Adobe already has done.  The Auto button works far better than any crappy cell phone processing.

Ten parameters with one click.

 Reilly Diefenbach's gear list:Reilly Diefenbach's gear list
Nikon D500 Nikon D850
robgendreau Veteran Member • Posts: 6,236
Re: Computational Photography

chipmaster wrote:

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

Not sure what you're saying. Google and Apple and Huawei are into computational photography, yes, but are the opposite of "marginal" companies. Nikon and Canon combined are lunch money by comparison.

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

Again, more like small boys. And that goes for the computers, which are at the heart of the phones and the, well, computational part. An iPhone now has a processor, eg, that is in many ways faster than what many of us have on our desks. So being able to do almost real time stacking for HDR, astro, focus, etc is more feasible on a phone than certainly in most any camera out there today. Not sure what an A13 chip costs, but I imagine it would boost the cost of a camera considerably. Hence a good reason not to include that, esp since most of use would do the computation on our computers (or tablets or phone) anyway.

Now the laggards have for the most part beaten the incumbents for total % of pictures taken by the majority.

That happened probably within a month of the first iPhone camera, computational computing notwithstanding.

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

https://www.nytimes.com/2019/10/15/technology/personaltech/google-pixel-photography.html

I'd love to see better computers in camera bodies. But not for stuff I do in Ps or Lr anyway. More for the bursts and AF and other features that phones are using. I'd still want the heavy computing to be more manually controlled in post.

-- hide signature --

“Art is not what you see, but what you make others see.”
— Edgar Degas

 robgendreau's gear list:robgendreau's gear list
Pentax K-1 II Pentax FA* 50mm F1.4 SDM AW
Jake2046
Jake2046 Regular Member • Posts: 242
Re: Computational Photography

why would camera companies need computational when their picture quality is way better than smartphone camera? What they need to do is add social media features...so the picture can be shared instantly.

Richrf
Richrf Contributing Member • Posts: 675
Re: Computational Photography
1

It seems like all digital photography is computational by necessity. The goal is to try to create an image of a continuous world utilizing discrete dots. For the most part the results are abysmal which is why there is such a reliance on the post processing correction process.

Current ILC camera manufacturers are burden proprietary technology and software and simply do not have the resources to compete in the future evolution of in-body computational digitatation which is what the vast majority of consumers are looking for.

Better glass and bigger sensors are always an advantage but not if the camera software is producing poor sooc photos and puts the reliance on external computer software. Current manufacturers will continue to decline in sales and more importantly in usage and I see no b way that they can reverse the tends.

Lee Jay Forum Pro • Posts: 55,002
Re: Computational Photography

Jake2046 wrote:

why would camera companies need computational when their picture quality is way better than smartphone camera? What they need to do is add social media features...so the picture can be shared instantly.

Unless you're a journalist, I can't imagine a reason to share an image instantly.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
Jake2046
Jake2046 Regular Member • Posts: 242
Re: Computational Photography

Lee Jay wrote:

Jake2046 wrote:

why would camera companies need computational when their picture quality is way better than smartphone camera? What they need to do is add social media features...so the picture can be shared instantly.

Unless you're a journalist, I can't imagine a reason to share an image instantly.

Smartphone cameras only popular cause photos can be shared with ease among friends and families.

robgendreau Veteran Member • Posts: 6,236
Re: Computational Photography
1

Jake2046 wrote:

why would camera companies need computational when their picture quality is way better than smartphone camera? What they need to do is add social media features...so the picture can be shared instantly.

Why? Because computational photography is used even with high quality cameras by some of us.

Examples are in-camera HDR, focus stacking, stacking for noise reduction, pixel shift, and so on. I prefer to do this in post, but sometimes it's nice to do it in-camera as well. I'd like to see it for panoramas for example, so that I could preview results rather than waiting until I return home.

But I agree about better sharing. I was asked on a recent trip to share and given I shoot raw it was a too convoluted procedure to get them an image compared to my phone.

-- hide signature --

“Art is not what you see, but what you make others see.”
— Edgar Degas

 robgendreau's gear list:robgendreau's gear list
Pentax K-1 II Pentax FA* 50mm F1.4 SDM AW
Lee Jay Forum Pro • Posts: 55,002
Re: Computational Photography

Jake2046 wrote:

Lee Jay wrote:

Jake2046 wrote:

why would camera companies need computational when their picture quality is way better than smartphone camera? What they need to do is add social media features...so the picture can be shared instantly.

Unless you're a journalist, I can't imagine a reason to share an image instantly.

Smartphone cameras only popular cause photos can be shared with ease among friends and families.

I thought they were popular because you already bought the thing for other purposes and always have it with you for communication.  So the cameras are just the lowest-common-denominator.

In the 7 or so years I've had a smartphone, I've shared one image instantly, and that one was unnecessary.

-- hide signature --

Lee Jay

 Lee Jay's gear list:Lee Jay's gear list
Canon IXUS 310 HS Canon PowerShot SX50 HS Canon EOS 7D Mark II Canon EF 70-200mm F2.8L IS II USM Canon EF 8-15mm f/4L Fisheye USM +23 more
PhotoFactor Veteran Member • Posts: 3,227
Re: Computational Photography
1

I'm thinking Sony with its electronics background is more likely to be early on computational capabilities approaching that of the mobile phones.

Perhaps Panasonic, too.

I just don't see Nikon or Canon getting too far while they are trying to scramble to get mirrorless bodies and lenses and catch up to Sony and others.

 PhotoFactor's gear list:PhotoFactor's gear list
Samyang 12mm F2.0 NCS CS Sony a6000 Olympus OM-D E-M10 II Sony E 18-55mm F3.5-5.6 OSS Sony E 50mm F1.8 OSS +5 more
PhotoFactor Veteran Member • Posts: 3,227
Re: Computational Photography

PJPfeiffer wrote:

chipmaster wrote:

An interesting article on the new phones., after all this site is about digital photography.

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

Now the laggards have for the most part beaten the incumbents for total % of pictures taken by the majority.

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

https://www.nytimes.com/2019/10/15/technology/personaltech/google-pixel-photography.html

I wonder if a similar discussion took place about Kodak when digital entered photography?

Here is an interesting side anecdote.

Around 2015 I worked with a guy who had previously worked at Kodak in Rochester, NY for many years.

He told me that in the early 2000's (2001?), he attended a meeting at Kodak. They saw the early emergence of digital photography. They discussed the potential threat to their film business. But they felt that they had 20 years to address that problem before digital would probably overtake film.

In 2012, Kodak went bankrupt.

 PhotoFactor's gear list:PhotoFactor's gear list
Samyang 12mm F2.0 NCS CS Sony a6000 Olympus OM-D E-M10 II Sony E 18-55mm F3.5-5.6 OSS Sony E 50mm F1.8 OSS +5 more
PhotoFactor Veteran Member • Posts: 3,227
Re: Computational Photography

j3ffw wrote:

chipmaster wrote:

Now why is the thrust being led by the most marginal of companies with the smallest sensors, because they were so behind and necessity is the mother of all invention.

or a lack of info (small sensor) and an output jpg that could do the processing in a timely manner

Why the traditional big boys didn't do it, because they were so preoccupied with their market share and competing against their decades old rivals and a classic case of innovator's dilemma.

The article states hdr that wasn’t an phone camera exclusive one could also say focus stacking, pixel shift, a stitched panoramic file. So to jump ahead, me personally I want the raw file, what will be nice is more advanced apps that one can program to set shooting sequences, iows an app that can run shooting scripts through the camera. You are right why have the manufacturers not added more sequences than just hdr and pixel shift. When will hdr pixel shift and focus stacking work in tandem? I’m sure the answer is once you code the software for it to work

For one thing are the software companies able to handle it? Capture one can’t do pixel shift, not great at panorama stitching or even hdr combinations. While you can’t really blame the camera manufacturer there is no doubt all of their apps are lacking

Yeah press/sports/wedding/nature/hobby they cling to the old, but once the new leaders figure out zoom and low light action who knows.

this is a funny statement because I have to ask, do you even know? What is zoom, what is low light action?

So let's not debate the phones and their quality, what I want to know from you all which of the big old companies will lead in the computational ? The laggards or the big three, I'd say the laggards likely have the most to gain to try, but do they have the resources, or do they partner with a google, or could any of the big three swallow their pride and approach google

what has google done? I don’t want a heavily processed jpg that can only hold up to viewing on a 6” phone with a limited attention span.

Have you owned a Google phone? I have, for 5 years. I don't think you really understand where we are unless you have some real experience.

 PhotoFactor's gear list:PhotoFactor's gear list
Samyang 12mm F2.0 NCS CS Sony a6000 Olympus OM-D E-M10 II Sony E 18-55mm F3.5-5.6 OSS Sony E 50mm F1.8 OSS +5 more
PhotoFactor Veteran Member • Posts: 3,227
Re: Computational Photography

Reilly Diefenbach wrote:

what has google done? I don’t want a heavily processed jpg that can only hold up to viewing on a 6” phone with a limited attention span. Seems like the next major step will be when Capture One or “that which shall not be named” add these features to their software.

Adobe already has done. The Auto button works far better than any crappy cell phone processing.

Ten parameters with one click.

Assuming that's true (I have some doubts), that's great.

Take a picture.

Copy it to computer.

Open LR, find photo.

Hit auto button.

Save file.

Then share or print it.

Lots of steps there.

Multiply by 200 or more for a photo session.

 PhotoFactor's gear list:PhotoFactor's gear list
Samyang 12mm F2.0 NCS CS Sony a6000 Olympus OM-D E-M10 II Sony E 18-55mm F3.5-5.6 OSS Sony E 50mm F1.8 OSS +5 more
Keyboard shortcuts:
FForum MMy threads