Hacker News new | past | comments | ask | show | jobs | submit login
Cameras are about to get a lot smaller (economist.com)
193 points by spuz on July 8, 2017 | hide | past | favorite | 86 comments



> He was holding a small device in his hand, the size and shape of a lollipop.

"This is a video camera, and this is the precise model that's getting this incredible image quality. Image quality that holds up to this kind of magnification. So that's the first great thing. We can now get high-def-quality resolution in a camera the size of a thumb."

...

"But for now, let's go back to the places in the world where we most need transparency and so rarely have it. Here's a medley of locations around the world where we've placed cameras. Now imagine the impact these cameras would have had in the past, and will have in the future, if similar events transpire. Here's fifty cameras in Tiananmen Square."

...

"There needs to be accountability. Tyrants can no longer hide. There needs to be, and will be, documentation and accountability, and we need to bear witness."

...

ALL THAT HAPPENS MUST BE KNOWN

-From "The Circle" by David Eggers


What good is video evidence, if it can be faked? Having numerous independent feeds helps, I suppose. But once you have enough perspectives to create a decent 3D model, you can generate as many fake perspectives as you like.

Also, there are obvious downsides to panopticons.

But none of that matters, I guess. We will have the cameras and the fakes.


>What good is video evidence, if it can be faked?

Even better: what good is video evidence if few care?

Consider officers shooting black people. There are several videos of that, but does it show any sign of stopping? They even walk off without any repercussions at trial...

And, conversely, when people do care, no extra evidence is needed.


Well I guess what people really care about is the story. The narrative that you derive from the pictures (and assumedly, audio recordings). And a TLDR of that. That would be an automation problem.

A dozen high-quality TLDRs and you've got the local, state, national and global scoop, as well as the latest dirt on politicians and movie stars.

I think that people would gobble that eagerly.

So a million little cameras = automated news.


Nearly everyone cares. But the people you're referring to carefully weigh all the facts (what little there usually are) before coming to a conclusion.

In which trial was it unambiguous that the police broke the law and also demonstrated institutionalized racism, and faced no repercussions?

I'm all for police accountability, but when we do have video evidence it often ambiguous on if the officer was in the wrong in using force. And when it isn't ambiguous, the officer goes to jail.


Actually people do care, especially since there is video. what they still don't care about that would probably need video as well is the staggering amount of black on black shooting.

however as a society are we willing to give up that much privacy the moment we step out the door, where anything and everything could be a recording device. then again if we are drowning in being observed do we in turn find more freedom?


With as many cameras as we already have, there isn't much privacy anywhere, unless we live alone.

I just reread The Light of Other Days by Arthur C. Clarke and Stephen Baxter. Basically, they argue that group consciousness is the best option where there is no privacy.


Right, sadly enough. Plus the fact that people will believe what they like, and likely won't trust all that complicated technical stuff.


Truth is subjective, culture, etc...

Eventually the science fanatics (empiricists) will realize this or get defunded of their jobs.


There is no "truth". There is only more-or-less likely untruth.


What good is video evidence, if it can be faked?

It's easy to "fake" news, and has been since cameras were invented - just choose the way you frame the image, or the moment you press record. No-one will doubt the video evidence of police beating protestors, and no-one will believe the claims that just a few minutes earlier they were throwing petrol bombs, after all where's the video?


It cuts the other way, too - with a sufficiently pervasive propaganda apparatus, it's easy to frame dissent as unpatriotic, as sedition backed by foreign governments or even as terrorism. If you've been told ten thousand times that enemies of the people are trying to undermine the state, evidence of human rights abuses or corruption might not seem particularly damning.


Not on the police cameras, but it still happened, they swear!


Of the fakes: Some NGO:s are vetting film material to tag specific ones as credible evidence of e.g. states behaving badly. They compare time of day to known weather patterns, cross reference previous film to see if something has been reused etc.


You might be referencing Amnesty International. They have create a site and tools that citizens and researchers can use to verify media as evidence. [1]

[1] https://citizenevidence.org/


Thanks, that I was.


That's a good idea, for sure. But all that stuff can be faked too. And the most resourceful players will have the best fakes.


A possible solution to some of the problems would be creating (and for official videos, requiring) a chain of trust, enforced by strong crypto, starting at the original recording device and ending at the viewer's display. Some trusted parties would also probably have to vouch for the start of the chain, too.

That basically sounds like the dream world of DRM people, though.


DRM I don't like. But authentication to reporter, recording device, etc would certainly be useful. Plus proof of integrity.

Maybe also cross-linking chains of trust with nearby devices. So devices corroborate each other.


There's an analog hole here ... what's to stop me using a trusted camera/chain/everything to take picture of an 8K screen with a fake image?

(Canon had a "trusted" model as such. It was broken, IIRC by elcomsoft, shortly after release).


Or a Potemkin village.


A camera that shoots 360°, saves coordinates and date, would make it harder.


How could it tell coordinates in either space or time? non-military-grade GPS is spoofable (not really easy yet, but easier by the day) and GPS signal is rarely available inside a building (and sometimes just under an awning or cloudy days, or near skyscrapers - GPS in NYC is horrible; mobile phone GPS are as reliable as they are because they take cell towers and wifi into account).

The important point is " making it harder " - but how hard is hard enough depends on the party in question, and for many important ones, the simple answer is "not hard enough".


The people in a position of power are already in the best position possible to abuse a root of trust like that.

Look at the TLS x509 PKI, its a total complete mess.

This is not a workable solution.


What would definitely work is a blockchain. It would definitely make sense at least for reliably timestamping video recordings.


Only if you also publish your video (or a hash of it) at the same time.


Widespread high quality evidence will make the fakes costlier to produce. Even resourceful players have ultimately limited resources.

http://tvtropes.org/pmwiki/pmwiki.php/Main/DystopiaIsHard


I don't even think you need "the best fakes" these days - just have your press secretary flat out deny that something happened. If any reporters push back, stop asking them questions and/or block them from your briefings.

Works for a fair few countries.


>Of the fakes: Some NGO:s are vetting film material to tag specific ones as credible evidence of e.g. states behaving badly.

And who tags the NGOs? Of which a great many are sponsored from other states...


Ultimately it's the justice system that matters. You can have as many videos you want, if the legal forces dismiss them for "reasons" you can go moot back in your room. This just happened with cop shooting people.


See also The Light of Other Days by Baxter and Clarke, https://en.wikipedia.org/wiki/The_Light_of_Other_Days.


Yes, that's an eerie tale. I'm reminded of The Last Trumpet Project by Kevin MacArdry. Featuring retroactive immortality through wormhole brain scanning.[0]

0) https://anarplex.net/hosted/files/last_trumpet/LTP.pdf


If everyone has their camera rolling, it's impartial witness testimony, including the feed of the defendant.


"Tyrants can no longer hide."

Seems he missed the other side of the coin. We won't be able to hide from tyrants.


In this particular book, that angle is more of a minor plot thread. Those politicians who don't buy into using the necklace 'body cam' version of this tiny camera to become transparent (to become 'clear') to their 'constituents' begin to become ostracized or, worse, blackmailed and then jailed (e.g. ostracized). The plot does revolve around people judging the central character's every action in real-time through these tiny cameras.

'1984' is more along the lines of not being able to hide with the 'TV' that watches you while you watch it.


Or you missed who is the tyrant. Anyone who tries to hide is a tyrant.


"Beware of he who would deny you access to information, for in his heart he dreams himself your master." -- Sid Meier's Alpha Centauri


So, where are your bedroom videos?


Obvious /s is not obvious apparently?


Apparently. There are people that say the same things in earnest!


Yes it's a pretty scary future to imagine. I had just watched The Circle before posting this article and it occurred to me that story might be closer to reality than we realise.


We will probably end up in a world where anyone can be surveyed constantly. Minority Report probably was not far off. Once enough patterns of behaviour start to tricle into government databases eventually they will get enough data to identify patterns identifying various pathological people before they hurt themselves or others. Then the question is: what to do with the data.


Basically irrelevant to a technology for making cameras thinner by a factor of millimeters.



> Cameras are about to get a lot smaller

No, scientists have developed a prototype which can take fuzzy photos of barcodes.

They then go on to tell you what would be necessary to have their device equal a present day sensor in a phone, but they haven't made one yet.

In fact, no estimate is given for when this technology might be competitive with CMOS sensors. The article just points to his previous work as proof he can get some of his ideas to market.

Relevant XKCD: https://www.xkcd.com/678/

I am excited by advances in camera technology, but this headline is peddling research as a pending disruption to the industry, and I don't see any evidence of that in the article.


He's saying he can make a commercial sensor in 5 years. Now, that's admittedly not saying it'll be in phones in that time, but when it works it won't just be competitive with CMOS sensors, it'll do things they aren't capable of.


There are planar Fourier capture arrays, which are also lensless, can be implemented with existing CMOS process and seem to be better by every metric than this. AFAIK they are currently being evaluated for eye tracking sensors which aren't sensitive to resolution (128x128 is enough) but need to be flat.

There are also single-pixel compressed sensing cameras, which can be made lensless as well and are limited by the optical modulator specs.

The problem with lensless sensors is always the computing power required to reconstruct the final image.


Admittedly I'm reading between the lines here, but I suspect that these don't need computationally expensive reconstruction. The reconstruction is done entirely by setting up the delays correctly on the input sensors, then the image is formed by simple interference. You'd read the image off just like an ordinary ccd.


From the article:

"He concedes that there are challenges: improving the optical performance of the elements; suppressing spillover effects between different signals in the device; and honing the algorithms that calibrate the camera’s performance."

The line of "honing the algorithms" is the complexity orbital-decay is referencing. These types of sensors do need computationally expensive reconstruction to generate images we're used to seeing with traditional optics and sensors currently found in many consumer devices. The filtering and focusing work the lenses do still needs to happen. These sensors essentially rely on complex math to replace the finely ground glass.


Those algorithms are already very well understood from radar and sonar, though. It's not like he's starting from scratch.

And I'd take issue with the characterisation that they need "complex math to replace the finely ground glass" - what replaces the glass is the analogue photon detection, delay, and amplification channel on the front end. My suspicion is that the only "complex math" is done calculating the delays before capturing the image, not doing the reconstruction (again, unless I've missed something unique about moving from GHz to THz).


Five years? Read the aforelinked XKCD; it seems a pretty accurate representation of reality to me.


This is very much on the path I've seen in software defined radio (SDR). I am particularly interested to see when they invert atmospheric interference in real time. A telescope or spotting scope with a co-linear laser visible to the circuit should be sufficient. Then no shimmering heatwaves in the distance, just a clear picture.


Some ingredients for a always recording society...

- Super cheap and gigantic capacity storage.

- Wireless network like 5G.

- Wireless charging.

- Nano cameras.

I think there was a Black Mirror episode about this.


I believe that in some decades will have something like Google Street View, but in realtime, at least for major cities.


Many cities already have industrial cameras installed on the streets and inside public transport (buses, trams, subway), as a way to prevent crime.

Until recently the issue was that no one had technical and financial resources to analyze the data, except on court warrant, but once Big Data and computer vision technologies become cheap enough someone will definitely use that as real time surveillance tool.


Well there was white bear which was more of a philosophical debate and then there was Most Hated in the Nation which I think really hits home the development of small new technologies such as cameras and the consequences of keeping their power and security unchecked.


There was also The Entire History of You - where everything you see is recorded and can be played back at any time - leading us to hyper-focus on details from the past rather than experiencing the present.


He is talking about 'The Entire History of You'


Could this enable a light-field camera? [1]

[1] https://en.wikipedia.org/wiki/Light-field_camera


Sensor SIZE is still the primary determination of quality. Not seeing how this changed that.


When will this ready to replace my 35mm, full frame camera?

:D


He's talking about an optical phased array system. Once it's working, competing with a 35mm system "should" (famous last words) be a matter of scaling up the number of elements to match the lens size. The thing is, once that happens, it's got a huge advantage: it doesn't need any other lenses. At all. Any "lens" you could want is a software configuration away.


But is that true? What if I want a 50x zoom lens? How would software do what an entire assembly of precision lenses does?


It's not so much that software does what the lenses do, it's that software would configure the hardware to perform the same transfer function on the incoming photons. It's more a reconfiguration of the input than a processing of the output.

On edit: I should also point out that you don't get to escape fundamental optical limits with this stuff. You're still diffraction limited, so "lens" size still matters.


is there already a model on sale? Like lytro?


No, the technology demonstrated is only capable of imaging 8x8 pixels. Here's a link to the actual research: http://www.caltech.edu/news/ultra-thin-camera-creates-images...


Lytro doesn't quite work the same way. It doesn't capture the phase of the light, so it's a bit more limited.


I don't think I fully grasped the explanation, it's like some sort of slit-scan pinhole camera? Aren't those images going to be super noisy compared to lensed cameras with larger sensors?


On mobile, for me at least, the text content of this article disappears when the page has fully loaded. Anyone else? This isn't the first time I've seen this sort of thing happen


economist.com brilliantly bans you from accessing their site after you've clicked a few of their links. The text vanishes because they've sloppily hacked this limitation in client-side by checking a cookie after the page finishes loading. This leaves 99% of visitors thinking that the site is broken, so they leave without ever noticing the message about how they need to pay a ransom to unban themselves.

For sites like this, you can sometimes archive the URL to evade the ban.

http://archive.is/Ni7Lo


OT: I think it is a bit of a jump to equate subscribing to paying a ransom.


It's a bit like a ransom when they load the article at first, you start reading, and then the text disappears asking for payment. Fortunately, The Economist seems to just use cookies so you can still view the articles incognito.


I hate how supermarkets hold food ransom. Greedy


Escape the article limit garbage by using Private Browsing and the like.


RIP Privacy


It seems like economist uses Javascript to create the "You've reached your article limit" dialogue. Simple press ESC key to stop JS from executing on Chrome so that you can read the article.

Also here is my TL:DR summary of it if you're still trying to fight through the pay wall:

There is a thing called grating coupler that works like little high frequency antennas that receives light signals. When you put a whole array of them you will be able to do various scans of light signals to simulate the camera pointing at different direction, or fisheye, telephoto effects without the need of tilting or moving the surface of the array. The underlying computation relies on the ability to calculate and control the timing of signal travelled from each antennas, plus some classic signal interference and phasing issues. An 1cm x 1cm array will contain 1 million such couplers which would create a similar sized image as an iPhone 7 rear camera, but since there is no lens involved, the camera can be made a lot thinner.


Using the Brave browser with the block scripts option enabled seems to work, too.


Thank you! Brilliant Summary. Sounds like a pretty great invention


You can just open the link in incognito as well.


[flagged]


This comment has a strong blend of meta and ideological flamebait, which nearly always poisons the discussion. Please resist the temptation to do this, no matter what you're replying to.

We've detached this subthread from https://news.ycombinator.com/item?id=14729686.


:) :) Not even able to articulate why you think this is inappropriate! Especially on a thread that has been detached and hence nobody from the general public can see it.

The fact that you simply return to cowardly downvotting it shows that you're trying to compensate somehow for the lack of an ability to debate.

Bigly sad for you! :) :)


:)


I can see your point of protecting sensible people from having to argue their positions.

I would like to get your opinion on the parent to which my comment was made. I find the assertion that police kills blacks in a disproportionate manner to be also and even more inflammatory and provocative. I was commenting to the aspect of an unnecessary political comment with an inflammatory slant.

Proof is also the fact that the flagging mechanism was used as a censoring method. My post simply stated that he needs more proof for statements made.

I am very disappointed on the way this flagging process is used to silence valid and well argued counter points. It is not an intellectual discourse but just a points-based battle.


What a sad person you must be if you feel the need to downvote even this comment. It was a polite and with clear reasoning. The fact that you chose to downvote it shows pettiness on your side.

Quite a little man!


:)

I'm of the same opinion: you know you have no moral ground and no intelectual argument to make. You simply let yourself be driven by pettiness.

:)


>If this is what passes for substantive and documented talk on HN, then things are getting a turn for the worse.

You mean contentless comments that don't engage with the thread they respond to but instead lament some lost golden age?

>Is hn at an impasse point, where no more technical issues are up for discussion?

HN always had a mix of technical and non-technical threads (the major one being business issues, which are of course non technical -- it's a VC firm's creation after all). And that's because, despite what the name might suggest, HN is not about people with one-track minds.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: