935 Spy Stuff

You're listening to Tips from the Top Floor, episode 935 for July the 5th, 2023.

Hey and a warm welcome, it's Chris. You are listening to your favorite photography podcast, Tips from the Top Floor, coming to you from Hanover, from the viewfinder villa right outside the gates of Hanover in Germany.

I am happy to have you back and let's see, one, two, three, we're talking about Getty today, about Game Boy, Blade Runner, nerd level 10,000, something about science, National Geographic and some trend spotting.

Yes, there's a new trend in photography land, but let's kick this off with a slightly juicy piece of news from the photography world.

Renowned music photographer Alec Byrne has decided to take...

The Giant, Getty Images. This is about a copyright infringement lawsuit.

Now, Byrne is no small fry in the industry. He's known for his iconic shots of the London music scene from the 60s and 70s.

He captured legends like ABBA and Fleetwood Mac.

And what is this about? Well, he alleges that Getty Images has been stealing his photography pretty much.

Well, they have... Okay, so what he says, what the lawsuit says is that they've been selling his photographs without any agreement or contract with him.

He claims that at least 175 Getty Images customers downloaded and copied his photos through Getty's premium access subscription option.

And at least 62 customers purchased a la carte licenses to his images.

And he says, no, I did not give Getty any permission to do that.

And the lawsuit particularly focuses on a photo of ABBA that Byrne has shot in his private London studio back in 1974.

And he claims that he only discovered in 2021 that Getty Images were licensing this image out to their clients.

And we're talking like publications like the Los Angeles Times, Vox, DotDash.

They have published that image.

In particular.

And yeah.

So, I mean, interestingly, earlier this year, Getty Images itself, they sued Stable Diffusion as the AI image company.

Because Getty claims they had stolen over 12 million of its copyrighted photos.

And so it seems like Getty Images is caught a bit in the copyright conundrum.

Being both the plaintiff and the defendant in different cases around copyright.

It's a bit of a tangled thing here.

Yeah.

So, Getty Images.

Yeah.

I like them so much.

Cool project.

Here's an interesting project, which initially is like photography.

Well, okay.

So, have you had a Game Boy?

Chances are, yes.

Have you had a Game Boy camera?

Not that many people.

But, you know, this huge big blob of a thing that you would slot into the Game Boy cartridge slot.

And it had this ball-shaped big thing on the top, which is a camera that can tilt forwards and backwards.

And the images would be very pixelated.

But it's a style.

And it's a retro kind of thing that has always had its fans.

And so, here's...

Here's a modder called Chris Graves.

And he's given the Game Boy camera kind of a makeover.

Sleek, minimalistic.

It transformed that into a Game Boy mini camera the size of a Game Boy cartridge.

Now, that's not just a cosmetic change.

He's managed to condense the original Game Boy camera into this small cartridge size.

He's done this by using a custom.

A custom circuit board.

And a smaller lens.

Actually, a lens from an iPhone XR.

But the mini camera uses the same sensor, the same chipset, the same ROM as the original.

So, it's pretty much the same camera.

And I've seen comparison images.

Yeah, pretty much the same.

But it is the size of a Game Boy cartridge.

And it looks like one.

Just a little lens on it.

And the shell of that camera is 3D printed.

It opens up possibilities.

It's a pretty cool project.

So, if you're a fan of the Game Boy camera, this might be right up your alley.

Okay, this next one.

Here's some spy stuff.

You know, there's things in photos that might reveal information about you that you might not want to reveal.

Here's a good example.

Keys.

Classic example.

Has been possible for a long time.

Take a photo of a key.

Or someone dangles their car keys into the camera.

And now people can recreate duplicates of those keys based on that photo.

Or look at even fingerprints on a photo.

With a current camera with decent resolution.

Fingerprints.

You could use fingerprints of photos to make like a fake finger to trick a keypad into letting you in.

Cameras are becoming increasingly...

Very high resolution.

And that's awesome on the one side and a little bit scary on the other side.

Now, enter a Blade Runner-like thing that...

Yeah, this is really interesting.

A research team from the University of Maryland has developed an AI-powered method that can reconstruct complex scenes and objects in 3D.

Using only the reflections in a person's eye.

Yes, you heard that right.

The reflections in your eye.

So, our eyes do reflect the room around us in one way or another.

Acts a bit like a wide-angle lens.

So, you get a distorted version of that.

And that team has overcome a lot of challenges.

Such as they have to compensate for not just the shape of the iris but the texture of the iris.

That...

Gives you like a big...

Like a backdrop that is not easy to separate from what's in front of it.

And then...

They pulled that off.

And they did real-world experiments.

So, they were able to successfully reconstruct a room in 3D using these eye reflections.

And that is something straight out of a sci-fi movie.

This is very Blade Runner.

So, they...

They used NERFs, Neural Radiance Fields, for that.

Which, again, can use...

Can create 3D scenes from 2D data inputs in some way.

Now, if you see the examples, yes, you do not get like a high resolution of what is around it.

But they had stuff on a table in front of that person.

And that stuff was 3D-ified based on the reflections in their eyes.

So, this is fascinating.

And it's also a reminder of like how technology and science continues to push the boundaries of what's possible.

It could well be...

By the way, there's also AI in that, of course.

And this could well be something that in future adds information to AI processing of images.

And might make photography different and better.

But, of course, it raises interesting questions.

What does this mean for privacy?

Like, could this technology be misused?

Can we...

If we want to...

If we want to shield ourselves from people prying into our private life,

should we wear sunglasses from now on?

Always?

Or special lenses?

Special contacts that...

I don't know.

Reflect something else?

I don't know.

It's almost the future for photography.

Photography is a moving target.

And this is...

It's our job to keep up with it.

So, let's keep our eyes open.

Pun intended.

And see where the stick says.

I used to...

Yeah, this is weird.

This is really weird.

Next one.

Nerd Level 10,000.

So, it's a fascinating project about photography and how it like physically works.

Like, you know, there's a light ray and it goes through a lens.

And that lens bends the light.

And it...

Then goes through an aperture and through more lenses and then hits a sensor.

And it gets captured by either red, green or a blue pixel.

And then...

And so on and so on.

Enter Blender.

Blender is an open source 3D software that is used to render movies and stills and things.

It's been used in some big productions.

It's pretty much state of the art.

And it's free and it's awesome.

And Blender has become so good that it can do...

It can do physical simulations of light.

So, it's always...

Like, there's always modules in there that can do physics in terms of like things bumping into each other.

And to make that realistic.

That's very important to make a realistic movie.

But this goes down to the level of simulating light.

So, a photographer...

This is a link to a video in the show notes.

If you're nerdy enough, you might want to watch it.

A photographer has taken that to the extreme and he has built a camera in Blender that simulates light in the optics.

So, it's like he does...

He makes a model in Blender of a box with a hole.

Makes a pinhole camera.

And then he cranks up the amount of like samples for that operation.

And that makes a picture.

It simulates the light going through that hole.

And on the other side, he places like a canvas and it makes a picture.

A pinhole picture.

And he continues to build a proper camera box.

And then he builds a lens.

Like a glass lens made of virtual glass.

But that simulates how light gets bent in glass.

And so, he has a lens that then projects the image into the camera.

And then he builds...

He continues.

He builds virtual film.

As in like three layers.

That capture different wavelengths.

And then he combines those three back into a color photo.

And then he builds even like an aperture into that lens.

That can be changed by turning like a virtual aperture ring.

You can link these things together physically in Blender.

Now, it's not 100% perfect.

But the photos are amazing that come out of this.

And even Blender...

Limitations, of course.

But it's damn near close.

And this is...

Really, this is nerd level 10,000.

This is so awesome.

And the resulting images are great.

The article in the links...

Or the video in the links shows some of them.

And...

Down to the bokeh.

And other kind of artifacts generated by those virtual glass lenses.

Now, again, not perfect.

He writes under the video.

Let me quote him.

Try this at your own risk.

I had this idea.

10 years ago.

Finally got around to it.

Took a good couple of months too.

Maybe I'll have to come back in another 10 years

in order to properly simulate diffraction and lens flare.

So, yeah.

Highest recommendation.

Go check it out.

Wonderful photography-related science-y nerd stuff.

Speaking of science.

There's an interesting field in scientific photography.

Art conservation.

Let's make a jump to art conservation.

So, if you look at old paintings, they have varnish on them.

Patina.

Like the gunk of hundreds of years that has collected on those images.

And this is usually a very weird brownish kind of layer.

That makes the colors duller.

And that people who want to conserve these images and restore them to their original pristine colors,

they have to take that varnish off.

It's one part of it.

And it's a painstaking process.

Because you'd see people with cotton swabs and cleaning solutions

and trying to guess.

Where does the varnish end and where does the picture start?

So, you only take the varnish off and not dissolve part of the actual color.

So, a team of scientists at King's College in London

has developed a super sensitive camera

that's about to revolutionize the way that we preserve our cultural heritage.

It's a camera.

It's not your run-of-the-mill DSLR.

It's a camera that uses...

It's like a new-ish imaging technology.

It's called macroscopic fluorescence lifetime imaging.

FLIM, for short.

And FLIM isn't new.

It's been used in the medical field to study and to track cancer cells.

And it even detects brain tumors in these things.

Very clearly delineates tumors from the rest of the surrounding stuff.

But...

This time, it's being used for art conservation.

And that might be a bit of a game-changer here

because, again, traditionally, art conservationists,

they have had to, like, really know what they're doing.

They would use ultraviolet light and a whole lot of guesswork

to get it right

and really relies on the skill of the person.

And it's prone to not being entirely accurate.

But this new camera system,

removes the guesswork by picking up, like,

the minute differences in how this centuries-old varnish

appears next to the paint and the canvas.

So, the system has...

I don't know.

Technically, I don't fully understand it.

But it kind of makes the...

Like, you shine a blue light against it

and then the camera kind of times,

every pixel times how long it takes,

like, the varnish fluorescing.

And that makes it possible.

For some...

In some way or another,

to really exactly tell those apart,

make them visible.

So, yeah.

Art conservation and photography.

All right.

National Geographic.

National Geographic has a very special place in my heart.

The magazine.

Because, I mean, if you were in the United States,

this has always been around for you.

But, you know,

a German boy who grew up in the 70s and 80s,

that was different.

We didn't have National Geographic here.

It was an American thing.

For the longest time,

they started bringing it over here later.

But back in that time,

we didn't have that.

But,

our family had an American friend from California,

Valentina Valena,

the old lady.

She's not alive anymore,

but she...

Yeah, I remember her visiting

back in the 80s a few times.

And when I was around 14, probably,

for my birthday,

she gave me a subscription to National Geographic.

And for two to three years during that time in my youth,

every month I would receive an envelope

from the United States

with the new issue of National Geographic.

And that helped

pique my interest

in the world,

in traveling,

and in the English language.

Because if I wanted to read something,

I had to translate it.

And that's why this next story

kind of really touches me on a personal level.

Because it seems like

the winds of change are blowing

through the hallowed halls

of National Geographic.

The magazine is known for

its amazing photogénesis.

Journalism and wildlife photography.

They have reportedly laid off

the last of their staff writers,

along with a number of other employees.

Well, okay.

So this is according to the Washington Post reporting.

According to them,

19 employees, including staff writers

and members of the audio team,

I guess,

around...

Well, you need audio production these days.

They already were given

the bad news.

The bad news back in April.

And this kind of coincides

with the widespread layoffs

of their parent company,

which is Disney.

Now, there is a bit of a mixed message here.

Because while it's been reported

that all staff writers have been let go,

there's an internal source somewhere

that suggests that the magazine

still employs writers and editors.

So there's a bit of interpretation

here. This could mean that

dedicated writing roles have been eliminated.

Maybe article assignments

are now being outsourced

to freelancers

or cobbled together

by editors or written by...

I have no idea.

There's two wildlife crime reporters

who remain on staff.

Their salaries are apparently funded by

the Wildlife Watch Program,

which is supported by the non-profit

National Geographic Society.

Um...

They already had some layoffs back in September

2022, like the six top

editors were let go

back then.

And they claim they will continue to publish

monthly,

but the copies will no longer

be on US

newsstands.

So...

Yeah.

It kind of... It really hurts, right?

I grew up with National Geographic

and it seems to be

pulling back on the legendary photography

that it's known for.

The magazine has reportedly

curtailed photo contracts that allowed

photographers to spend, like, months

in the field. Yeah, and National Geographic

insists that these staff

changes won't impact

the quality of their storytelling.

Um... Yeah.

Just a bit

of history here. Disney purchased National Geographic

back in 2017

as a part of a larger merge with

21st Century Fox.

And that...

That was kind of a turning point

because that's when the magazine turned

into a for-profit

organization.

Up to then, it was

a non-profit journal of

the National Geographic Society

since 1888.

So, there you go.

Alright, and last but not least,

this...

This one... I thought this was really weird

initially, but then...

Yeah, okay, so here's the story.

Influencer Liliana Madrigal

has stirred up

a bit of a buzz on TikTok.

She claims that the

secret to the perfect selfie isn't

your iPhone's camera, but rather

the screenshot function.

Yep.

They called it screenshotting. So,

um, I would

have shrugged this off, but

that video has racked up, like,

4 million views

at this point, I guess, roughly.

And, uh...

Uh, the...

What's her name again? Madrigal?

She takes a jab at the iPhone's

camera quality for selfies.

Calling it really bad, and

um...

She advocates for a pretty

unconventional method, posing

for a selfie on the front-facing

camera, and then screenshotting the image

from the camera app on

the screen. And then she

crops it and

makes it look like a regular photo

without the screen elements.

And it's notable that this,

this thing received, like,

400,000 likes

and, uh,

lots of comments

calling her genius for that tip.

And, of course, I mean,

this reminds me of when I was

a kid, I...

I grew up, again, in the 70s,

80s, um, back then, I

I record, sometimes recorded

something onto a cassette recorder

by just putting the cassette recorder

in front of, let's say, a TV and

record through the microphone,

so very unconventional, but

it did the job. It

wasn't good in quality and anything,

but it did the job, and it was okay, and

for the time, it was plenty good.

Now, of course, we, as photographers,

we have these cameras with

5,000 megapixels and

super quality and so on.

And, uh, so I tested this.

I tested this, and of course,

of course, resolution is one issue.

12 megapixels,

that's what the front camera does, versus, like,

the 2 megapixels of the screen.

So you get this, um,

drop-in resolution, which

is part of the image, like,

the image comes out softer, of course,

at the same size, so, um,

it is really good at hiding

blemishes in these things, so that's

one of the effects that I'm pretty sure we're seeing

here. Um,

the other is

that it's not mirrored.

Like,

if you do that, then

you see yourself like you see yourself

on the camera, which is mirrored,

um, and that

might change the perception here.

And then, last but not least,

the one thing that this live preview on the screen

doesn't do is all the processing that

the iPhone camera does, and

if you've ever looked into this, the iPhone camera

takes photos all the time without you pressing a button,

and then when you press the button, it uses

photos taken

prior to your pressing the button

and after you're pressing the button. It's like

5, 10, 15, I don't know how many

photos that will then be

fed through the, the,

uh,

optimizing algorithms and

bring out the detail

and so on, and it's, it's a different

look than the, just

a plain photo from the camera that

is, it's

also processed, but not to

that extent, and

a lot of people seem

to prefer the

air quotes crappy

kind of quality from a screenshot

to what the camera tickles

out, so, um,

I think it's just, it's just

um, again

as in air

quotes, a serious photographer

that made me chuckle, but

then I'm also

witnessing a new generation

dealing with photography

completely non-technical, completely

they don't care if

it's 2 megapixels or 12 megapixels

because it will be used in a context

that doesn't really

rely on resolution

um, the processing

isn't, isn't working, so they're

they do it in a different way, and it's

uh, it's fine, it's

good enough, it doesn't matter

if it's not perfect

so, they, they experiment

um, and

I think that's a refreshing

approach to photography, even though

I probably wouldn't do that, but

yeah, screenshotting, it's apparently

a thing, so

there you go

and that was it for today's episode

thanks for listening, thanks for subscribing

thanks for leaving a rating or a review

wherever you get your podcasts

and uh, as always

you can leave feedback for the show at

tfttf.com

slash hi

let me know, do you have questions?

do you want me to discuss a specific topic?

do you have an anecdote or something

that happened to you that you

uh, that you are

curious about? let me know

tfttf.com

slash hi

here's Mark, who has done just that

he writes, hi Chris, we

are certainly living in another photography

renaissance, or the death of it

I listened to a Zoom call with ASMP

yesterday, and it's really the wild west

out there regarding AI, a lot has

to be sorted out on both

importing of images and their export

or usage, Adobe is taking the right

path, but I feel that Pandora's

box has been opened, and it will

be ten times bigger in a week

regarding using AI for street photography

I think it's a losing battle

you have to remember that AI doesn't make

a face, I could be wrong

it is taking a photo

from a source and replacing it

I believe this will all

come to a head when faces can be tracked

eventually, I just think

about how that

will affect street photographers, once

someone finds their face in your photo

the lawyers are going to have a field

day, now add to that an AI generated

face of someone else, who the

AI didn't get permission from, and you don't

get permission from, what happens

if the face goes onto

a body that everyone recognizes

when can it be considered art

and can it be printed in

a book of street photography or

sold as a fine art print for your wall

I love the street photography genre

but I fear it may be

at its end once people can

find your photo of

their face in the image, thanks for the

great work you do. Hey Mark

long time no hear, glad

you're still in

photography professionally

this feedback goes back to the

article

about the photographer replacing people's

faces in street photography

using AI generated

different faces, I think

I believe it was in the last episode

I agree that this is a bit

of the wild west right now

and that the pace of change is

kind of frightening

but there's one important thing to understand

and that is the AI does not

just copy and paste

it generates

so these generated faces are

like they're really generated

but of course

based on training on real world

material, on photography

out there, on people's faces

it's

the best comparison I can come up with

it's a bit like, let's say you study

painting and you study

what a Picasso looks like

do this really well, spend

years and then you try to paint your own

Picasso, what you paint

might look like a Picasso

it might even come close but

it's not an exact copy

it's a newly generated

piece of art that you just made

based on what you learned

and the AI does something similar to that

all those current lawsuits

that target AI companies

are not because an AI would make

an exact copy or make a collage

of copyrighted snippets

that is not how it works

those lawsuits are about

the material that the AI uses

for training

so Getty for example, suing

Stability AI, that's what they

are suing them over

that Stability used

Getty's images

for training

so we will see something

come out of that

I just don't know which direction that will go

and when it will happen

but yeah, let's

keep our eyes open

and again, if you want to

get your feedback here on the show

tftdf.com

and again, thanks to all of you who support the show on Patreon

Patreon is awesome

not just as a means of financial support

but again, it's a

it's an incentive boost for me

to bring out new episodes

because there's so many things

tucking at my calendar right now

so yeah, thanks of all who support

your favorite creators

not just this show

but there's plenty more out there

who use Patreon

you can find out more at tfttf.com

slash support

that's tfttf.com

slash support

and as usual, Patreon supporters are first in line

you get to listen to new episodes

before anyone else does

so Eastern

European Electric Photo Road Trip

is a go

well, has been for a while

there's one spot open

one last spot open

on the way from Berlin

to Transylvania

find out more at

discoverthetopfloor.com

it starts on September 2nd

and now go out and take amazing photos

be nice to each other

again,

make sure to spend some time in the sun

it's summer outside

and of course

happy shooting