Spy camera detection using smartphone time-of-flight sensors
dl.acm.org
dl.acm.org
Google’s branding is awful.
Man, we're so bad at naming, that we keep joking about it... And trying to satire that with internal codenames (the ones engineers give without adult supervision)... And somehow those often seem better than the external names :(
This is something we'd have to separately test, and ultimately make a wholly different app for if it's feasible on iOS.
What would be really helpful is a database of apps, similar to the WINE project, maintaining how well they work on a de-googled phone, with a table, columns being "no root", "rooted", "with microg", etc, and a WINE style rating of how well it works and what problems you'll run in to.
You would know which banking services, etc don't work well or at all and could choose your establishments with foresight.
What were your biggest reliance on Google services you found in this 1 year and how did you overcome it?
However there is a clear bottleneck with the Google AR team for features that the community really wants. While it's understandable, perhaps the open source libraries aren't the priority or there's short staffing, there are issues like this one - https://github.com/google-ar/arcore-android-sdk/issues/153 - that have been open since Jan 2018. This one asks that we can use the smartphone flashlight simultaneously with ARCore, but there's no visibility on how close we are at the moment. This feature would have likely improved our work's performance greatly, but even in general, many other AR developers are asking for it.
Re: overcoming it, there's not much we can do in this case. We just didn't implement the feature.
Long standing issues/feature requests are typical of Google's Android ecosystem, Regardless I eagerly await for your release. I think ToF based spy-camera detection could make a great addition to the women safety apps in India.
You've got a community of people who're willing to spend time polishing.
It'd be a massive benefit to society to make this widely available.
Do you have a link to where you can buy these types of time of flight lasers/sensors? Curious about the additional hardware cost versus sensitivity.
While this work operates on ToF sensors that are already present in smartphones (e.g., Samsung S20+/Ultra), a Microsoft/Azure Kinect should also be a valid option because it has a depth camera as well. It has a higher resolution and bit-depth as well.
We initially intended to compare against the Kinect, but it doesn't fit the use case (something that you can have on you at all times). However, it could be a cheap choice for a different kind of deployment (automated hidden camera detection with robots, perhaps?)
They're basically for augmented reality applications because they sense depth. Placing objects at the right size and scale in the augmented view is much easier and more accurate with the ToF sensors, for instance.
Not all of them are lasers, which might be important.
what technology advancement would be needed to increase the detection rate and reduce the false positives?
- Increasing the resolution of ToF cameras (right now images are around 320 x 240) --> reflections from hidden cameras can then be more detailed, whereas now it's only 1 or 2 pixels each.
- Increasing the bit-depth of ToF images - right now every pixel is only 3 bits (8 colors). It's very hard to differentiate bright hidden camera reflections from everything else, so we had to do a lot of work for that.
- API improvements in conjunction with augmented reality libraries, e.g., a) allowing Android devs to enable the flashlight when AR apps are running b) more raw access to the ToF sensor if possible
Great project btw!
If you have a projector, you can do more. I believe Apple uses a flash, which has a low resolution, but is perhaps less cpu-intensive and less error prone, although it has a lower resolution. That would be a real Lidar, which is an active measurement. Of course combining that with sensible stereoscopy nets better results.
ToF stands for "time of flight". It works exactly by measuring how long the signal takes to go from camera to object and back [1]
Stereoscopic cameras are another type of 3R camera, they are not ToF and they are not active.
Each has pros and cons.
Most other ToF sensors use "indirect" ToF, where they measure the phase difference between incoming and outgoing signals to derive distance.
However, it gets murky as cheap 2D LIDARs on say, robot vacuum cleaners, use geometric techniques to find distance (basically return angle of a reflection). I explored this in a previous work.
TLDR: I would recommend not taking any naming at face value and reading the actual datasheet or more commonly, technical marketing materials, since few ToF manufacturers that I see have a public datasheet.
> many early "stealth" radar detectors were equipped with a radar-detector-detector-detector circuit, which shuts down the main radar receiver when the detector-detector's signal is sensed, thus preventing detection by such equipment.
https://en.m.wikipedia.org/wiki/Radar_detector#Radar_detecto...
As you say though, extra heat is a physical guarantee, so maybe a smarter technique exists to separate signal from noise in the thermal domain that I don't know of yet.
https://en.wikipedia.org/wiki/One-way_mirror#Principle_of_op...
Metamaterials have a chance at at: https://www.nist.gov/news-events/news/2014/07/new-nist-metam...
Counter-counter surveillance techniques, IIUC.
For the applications you suggest, there are some existing military-looking devices out there that use multiple lasers to find sniper scopes, for example. My basic searching shows at least https://www.ldsystems.us/product/sniper-optics-detector/#, though I'm sure there's more.
Regarding using the app to check, I guess that applies for the existing handheld detectors as well. It's definitely something that intelligent attackers can try to plan for, but we havent tested the adversarial robustness of the system right now. That would be a very interesting direction for us as well.
Single pixel cameras are lensless, for instance.
Here is a random one:
https://www.lexuma.com/products/lexuma-xscan-portable-hidden...
Anyone????
House Atredies?
I sense the Emperor behind it.
It's quite amazingly effective, especially the first time, when you're not really convinced.
Random google result:
I am not sure if it is worth the price, but it's cheaper than a top of the line smartphone with a time-of-flight sensor
- No automated detection, so lots of human subjectiveness on whether a bright spot should be investigated
- Background lighting can be an issue, so the view through the detector can be filled with bright reflections
- When I tried this for longer than 30 seconds, there's really some eye strain involved
The idea is that the smartphone has more than this one purpose, and the detection system can improve with software updates, whereas you get what you see with the detector and nothing more.
https://petapixel.com/2012/10/28/worlds-largest-private-yach...
https://www.kickstarter.com/projects/spyassociates/spyfinder...
> The same poll found that one in 10 guests (11%) had found hidden cameras in an Airbnb rental.
https://www.forbes.com/sites/suzannerowankelleher/2020/01/27...
Then you can make and sell your very own one using their effort.
The kickstater itself is a cheap knockoff that didn't bring any new feature or any new design elements. I was too lazy to do the research, but it wouldn't surprise me if they just rebranded a product offered on alibaba.
I would support them if they offered a superior design with a nice looking aluminum/Stainless/titanium case, made it more compact, or made it waterproof, etc. but they did nothing like that. They are offering the same cheap back plastic design that looks 95% like all the other cheap chinese knockoff.I think only the high end iPhones/iPad have these type of cameras right now right?
I'd also be curious about the exact angle you have to hit to get a reflection
Probably wouldn't work but bright flash in a room?
We also think using the smartphone flashlight (if that's what you mean) is the best way forward. That's already very helpful (and recommended) for humans to find hidden cameras, and it should be a useful extra modality for our work too.
I assume minimizing the aperture would be a fairly reliable way to avoid detection. A pinhole camera would be hard to find by any optical technique, though the video quality would suffer. Perhaps one could project through a pinhole onto a screen and record the projection, so the sensor is angled away from the room?
I think the easier way would be to hide the cameras in much harder-to-reach places so that it's inconvenient for the user to get their smartphone near. This might reduce the kind of videos that can be taken, but maybe an attacker will find that a reasonable tradeoff.
The spy camera could also something like your system to detect the phone camera and take defensive measures.
Installing an absorptive IR cut filter on top of the lens would decrease the amount of reflected light, and might hinder your approach. Those are pretty cheap to buy, so you could try it out pretty easily.
From this paper:
> Specifically, the hiddencamera embedded in the object reflects the incoming laser pulses at a higher intensity than its surroundings due to an effect called lens-sensor retro-reflection. This occurs when almost all light energy impacting an object is reflected directly back to the source (see Section 2.2). These unexpectedly high-intensity reflections from hidden cameras cause certain regions of the ToF sensor to be “saturated” and appear as black pixels. LAPD processes these saturated areas to automatically identify the hidden camera and its location and displays it on the user’s smartphone screen.
A stand-alone Single-Point LIDAR might be used to remove the high-end phone requirement. Should interface to a phone USB port with a simple UART/I2C-USB adapter.
Search Amazon.com for LIDAR finds 'Lidar Range Finder Sensor Module TF-Luna, Single-Point Micro Ranging Module 0.2 to 8m Compatible with Pixhawk, Arduino and Rasppbarry Pi with UART / I2C Communication Interface' $25.
And 'MakerFocus TFmini-s Micro Lidar Module 0.1-12M Lidar Range Finder Sensor Obstacle Avoidance Sensor Tiny Module 1000Hz Single Point UART I2C IO Compatible with Pixhawk Ar duino and Raspberry Pi' $39.
The 2nd one is waterproof. Also on Amazon 'Worldoor CC308+ Multi-Detector Full-Range All-Round Detector For Hidden Camera / IP Lens/ GMS BUG / RF Signal Detector Finder , CC308 + detector hidden mini camera/IP camera/general manager/radio frequency signal detector instrument' <$20.Camera array of a thinkpad T480 with windows hello (ir camera) and a hazard light/strobe (sourced... locally)
[1] https://play.google.com/store/apps/details?id=com.lvonasek.t...
Separately, Lubos, the creator of ToF Viewer, is very helpful to the ToF community and constantly answers any questions about APIs or sample code. I'd recommend talking to him if anyone wants to get started with ToF-based applications.
Looking at what the DL filter is doing, I wonder how well this really generalizes. It seems that many of those examples are virtually indistinguishable to a human (hence the high false positive rate for the naked eye experiments).
I also have similar reservations about the DL approach in general. What makes me more confident is that we trained on a relatively small number of objects (~10) and tested on a totally different set of 30 objects. We're also using very small regions-of-interest as inputs (5x5), so there's little to no overfitting on the environment or objects themselves.
I do this to test their batteries :)
But older Cell-phones (flip phones) with CMOS sensors used to be really good at this as they wouldn’t filter much IR in order to work better in low-light conditions.
However, if the camera reflection is always within a larger reflection for all possible distances and angles, then I doubt this will work.
If you want to spend some money and don't mind carrying an extra device: probably any of the "hidden camera detectors" on the market will be at least somewhat useful. K18, CC308+, etc.
We're trying with our work to get better or at least equivalent results without having to use external devices.
Basically, to keep position X unobservable through the hidden camera, ensure that there is an overwhelming source of photons between position X and the hidden camera.
Since we don't know where the hidden camera is, the best we cam do is to assume that the immediate vicinity of the bright light will accomplish this.
Thank goodness! Imagine being convicted of a crime because some perv took some pictures of your personal stuff in an attempt to frame you.
Contrary to popular believe, circumstantial evidence is not bad evidence. In isolation, it might not be enough to convict, but when used in conjunction with other evidence, it can create a damning case.
1. The camera contained pictures of owners desk.
2. That model camera was purchased on amazon by the owner.
3. The serial number of the camera indicates that it was sold on amazon and produced around the time of purchase by the owner.
4. The camera was found in a place the person had reasonable, unrestricted access to.
5. owner was found in possession of pictures that look to have been taken by the device, in the position where the device was originally discovered.
* I'm not asserting these facts are true, just stating them for the sake of example.
In isolation, each of these pieces of evidence don't prove much, but in totality, it is highly unlikely that all of those things would happen to an innocent person. Jury's don't need to be 100% certain to convict, they need a preponderance of evidence.
I can see why a prosecutor wouldn't pursue this case against a rich person though. The police are unlikely to do a good job at collecting evidence, a good lawyer will get enough of it thrown out, victims probably won't want to testify anyway, and being a business owner, this person might have clout with local politicians who will make trouble.
But you're right that's pretty damning.
IANAL, but since you refer to a prosecutor, they would need "beyond reasonable doubt", right?
It's a bit unnerving/cool that you'll see 10 pairs of eyes looking back from the lawn you might have just been lying on.
https://www.amazon.com/Headlight-Rechargeable-210%C2%B0Illum...
I'd be looking at something like that, at least as a first run over the room. But perhaps to augment. It allows you to see spider eyes at 5+ meters.
Most people will already know this, but LAPD = Los Angeles Police Department, often featured in TV shows, films and the news.
s/most people/most americans/
plus some non americans of course, but it's not a global association by any means.
With that said, while you're correct:
- A significant portion of tech advancement is US-centric, so I understand the instinct to ask for clarification
- I'd argue that the LAPD is relatively well-known world-wide, much like the RCMP, or RAF, or other very famous organizations