Boston Bombings - Outrageous Manufacturer Claims
As details emerge on the Boston bombing, the impact and use of surveillance technologies is coming into view. [Now, updated below with a list of manufacturer claims on how their products would have (ha ha) prevented or solved the bombings.]
The FBI says that one of the bombs was contained in a black backpack that looked like so:
Presumably, since no one suggests this is a suicide bombing, the bomber must have abandoned the bag outside.
Immediately, this raises the question of detecting the abandoned object, either through a person calling it in or an analytic system automatically detecting it. However, give the dense crowds and numbers of moving people, this would be very hard to do via computer.
The next question is the role of recorded video. Historically, this would have been whatever surveillance cameras were in place. However, a huge shift has occurred in the last 5 years. Now, literally everyone has cameras integrated into their phones and many people were recording images and videos before and during the incident. Given the sheer number of those cameras, their superior resolution to most deployed cameras, and the variety of angles they were taken at, those personal cameras could prove even more critical than public cameras.
There's been a debate about the 'crowdsourcing' of the investigation where people on the Internet are going through shared videos and images, looking for suspects. This image has gained the most attention:
This could simply be too guys hanging out, falsely accused on the Internet. Or it might be right, or there might be another lead identified by people on the Internet scanning through personal videos and images.
Questions: Any analytics that can help here? Either to detect beforehand or search? What about the role of traditional surveillance cameras vs. personal camera/phones?
To a point, I agree , but it is about ROI - NMW , and Exposure - NMW and they want the front page for exposure so as to get products in the limelight . Company Goals - NMW
Granted policies, goal's, objectives, public perception makes a difference, but what goes on at the management level (behind closed doors) is far from what you see in the public image.
I have sat in on too many meeting's to believe that a company does not operate at these extreme tactics to get in front of clients. They have trips, vacations, golf to accomplish these undisclosed ideals, special meetings
Microsoft is one of the best examples of how things really work. Companies know what it takes: cost, how to control damages, and Exposure. The bigger the client base the more tolerant they are of error.
I dont have that flexability. Larger Companies do (They can twist, turn, manipulate the truth) and get away with it.
Good Damage Control Policy's. Look at Tyco(ADT), Exxon Moble (BP), GE (Aviation)
They have all had great mistakes, blunders, and flat deceptive practices (They still survive well)
04/28/13 03:24pm
Bugs in software/firmware that hits the streets is one thing - and I will stipulate your given points above about development schedules/bad coding/marketing decisions, etc and the role these all play in all of us having to deal with non-perfect software. However, software/firmware bugs vs the actual efficacy of the product itself is apples vs doorknobs.
No company can stay in business (without periodic cash infusions) if they can not prove the fundamental value of their product. i.e. it has to be shown to actually do what they say it can. In the case of software/firmware bugs, the fundamental value is not in question; the technology has been proven to work.... even if not exactly all the time, every time (which is kind of what we all want/expect, no?) :)
In the case of some of these various analytic company claims, their fundamental value has never been proven (at least publicly), yet the massive media blitz that has been unleashed pushes hard to make everyone believe that their fundamental value is not even in question.
Christopher, sounds like Microsoft's model - get it out the door, fix it later.
Not that that's a recipe for success, there were plenty of other things that put Microsoft where they are today... come to think of it, most of them were kinda shady too... okay, let's not use Microsoft as a "good example".
@ Marty
So say : this may not be the case here, there is still some truth to it.
Years ago i did beta testing for a company called C&K out of Folsom Ca. , Well Known Company
This Company was really a chip manufactureer, But had a arm in the security industry.
They would knowingly put out the products so they could be in the front of industry , Make Corrections to the products , change thier idealogy in the process after the product was out .
Had some Great Products still great today ( Honeywell )
They knowing the products did not work properly sent them out the door , but because of deadlines,production,schedules ( UL ) , they would send out products that still needed work.
With the market where it is today and IT,Infrastructures,Internet capabilities so far along
It is very easy to upgrade, Make soft changes , and correct blatent mistakes on the fly to products as they are being installed .
We do soft flash updates constantly (eprom,eeeprom,softeeprom chipsets ) , which substantiates my claims of how things really work out thier .
Your company may not work this way , but some still do .
Bottom line is ROI, Invester returns, and Payback
I think the next-to-last paragraph of the story is almost as laughable:
"Many older cameras use proprietary formats, which cannot be read by analytic tools. “We have to make sure the systems we are putting out are in a standardized format where we can apply any type of forensic tool,” he said."
So is he saying that cell phones, still cameras and all surveillance systems should use the same image format? Good luck!
04/27/13 04:56pm
"That should be hung on a wall plaque somewhere."
John could probably pull in some extra coin selling IPVM T-shirts with that phrase on them.
04/27/13 04:34pm
"Seriously, if you do not understand why this is infeasible, you should not be anywhere near a security or surveillance system."
That should be hung on a wall plaque somewhere.
A government focused trade magazine site has an article with a highly misleading title: "How video analytics helps reconstruct Boston Marathon bombings"
Now, you might think that the article talks about what actually occurred but, no, it's a shallow speculation about what hypothetically could happen.
Manufacturers promoted - Briefcam and BRS Labs. Stupid claims include:
"So instead of spending 20 minutes looking at video in which nothing happens, the investigator can hit a button and in 30 seconds go to the area of interest and then begin to dissect what actually happened"
This is actually from an end user - the US Parks Police. Obviously, this 'benefit' is irrelevant in the case of the Boston bombing where there was fairly continuous movement all day.
The section on BRS Labs is simply incomprehensible. No attempt to explain how it would help with the Boston bombing, just that it scales to thousands of cameras because no rules are required.
Finally, they cite a forensic specialist who must be from a Tom Cruise movie:
"Using facial recognition software and video forensics, they then could cross reference security video of people buying pressure cookers with images — such as those of the two suspects — taken at the marathon finish line"
Seriously, if you do not understand why this is infeasible, you should not be anywhere near a security or surveillance system.
04/27/13 03:52pm
I could not disagree more with that thesis Christopher.
Truth is truth. Fabrication is fabrication. Truth is sustainable. Fabrication is not.
If the goal of your marketing is to fabricate stuff simply to 'get the numbers up' - even if this is a non-sustainable business model - they stop buying your stuff.
Once people figure out you are fabricating [especially if only after actually buying your stuff] they tend to tell everyone they know how they just got jacked by your marketing dept. Hence, my position that this is not a sustainable model.
So... what might be a potential motive for a company who 'just wants to get their numbers up'?
I think so many times in the industry, the companys just want quick fast exposure and they dont care what or how they get it.
Getting into the front of the industry can be positive or negative and have a good impact on the sales.
Presence is extreme in the industry.
If the marketing guys are correct and the demographics truly are right.
Then the companys listed above are right where they want to be: Front of the class, regardless of what the perception is.
It is a matter of Marketing, Not right, wrong, true, or false
They just want to get to the front of the line and be able to get the numbers up.
Exposure is everything good or bad.
I wont go undisclosed to say this .
After spending the last 2 weeks insinuating how wonderful face recognition was in identifying Boston bombers, now 3VR changes their tune:
"Facial surveillance is in productive use today, but was not likely much help for Boston. It takes a deliberate effort to set up facial surveillance and it is not well-suited for the ‘after the fact’ video being used, at least not yet."
"it's incredibly hard for any system to distinguish between a tall eight grader hanging out near a tree and an adult stalker."
...not to mention, distinguish between a short, gangly stalker and the normal group of kids.
End users need to share the blame in the desire for impossible surveillance solutions. In a new article on BRS Labs, the Texas Department of Public Safety statewide directory of security, and BRS customer, says:
“I needed something that was smart enough to learn, ‘This is a bus stop, this is where kids get off for tours, this is normal activity. It’s not normal activity if there’s a male 6 feet tall standing by a tree taking pictures of the kids.’”
He may need that but the likelihood that he is going to get that from any surveillance vendor is slim to none, because it's incredibly hard for any system to distinguish between a tall eight grader hanging out near a tree and an adult stalker.
Oh boy. According to the 'Today Show', cameras are now so high tech 'they can actually pick out a bomb even hidden on the street before it goes off'. The example given is another 'abandoned bag in an empty area':
One, that bag is obvious not 'hidden'. Two, why do these mock terrorists always place bombs in areas with no people? Do they want to blow up newspaper stands? Three, unfortunately, NYC is on this as they demonstrate the example.
The video concludes with a visual of the Boston bombings and BRS Labs saying they are changing the face of law enforcement and that 'catching these events before they happen is the name of the game.' Does anyone, in their right mind, think BRS Labs would have prevented the Boston bombings?
And here's another company claiming credit for solving the Boston Bombing, though this one may have legitimately been involved.
An Alabama TV station is reporting on how Intergraph's Video Analyst tool was used to help investigators piece together different video clips from multiple sources. Here's the TV report:
Matt, yes, I agree about that. It's certainly a lot easier to get non domain experts excited about such claims. That said, this is the physical security industry, so while some will certainly be a little more discriminating, I bet there will be enough that won't.
The reality is companies like 3VR or BRS Labs don't need 'everyone' to buy into their spin. If they can a dozen big wins a year, that's more than enough. PR Pays! :(
Guess it depends on how many actual DECISION MAKERS each has reached in their target audience, no? Sure, you can get Joe and Jane Average ooo'ing and aaaah'ing your name, but they're not the ones spec'ing your product. One hopes at least SOME of those signing the big checks for future projects are a little more discriminating than to make their choices based on media coverage.
HOPES, I say... trying not to be TOO cynical here. It's not easy...
Is it time to pull out this old gem yet? Sorry, I'm too lazy to photoshop in the appropriate players...
Well, if only life was so fair. Think about how many tens of thousand's of people 3VR's PR agency has duped and what a small number of people realize the reality of their posturing?
So, yes, certainly, those of us who carefully track these things respect Animetrics more than 3VR here but doesn't 3VR still win overall?
"Paul Schuepp, CEO of Animetrics ... doesn’t believe its products or other technology were the main tools used by authorities in Quantico, Va. to identify the bombers."
Wow, that's a refreshing bit of honesty! You'd normally expect an admission like "facial recognition did not identify Boston bomb suspects" to be followed up by something along the lines of "but if they'd used OUR system..."
See, kids, THIS is how you gain RESPECT, rather than just flooding your name through the media along with all the other wannabes.
Finally, a face recognition vendor admits the obvious - facial recognition did not identify Boston bomb suspects. The CEO of Animetrics concludes:
"You have this God awful picture of these guys and trying to take the photo and match it with one in a database ... I know they tried but I don’t think they got great results"
Good article at Boston Business Journal.
In Salon, there's an interesting interview with a Carnegie Mellon computer scientist about "Why Facial Recognition Failed," a solid review of the core technical barriers in doing so for the Boston bombers.
Here's an article in Gigaom about "Why Facial Recognition Software is Not Ready for Prime Time." In it, they quote the director of the National Biometric Test Center saying "video intelligence company 3VR’s products were not used to find the Boston bombing suspects."
IPVMU Certified | 04/24/13 11:58am
Mayvbe this should be a lesson to software vendors- you don't have to lie about your products. You can take an honest approach and protect your integrity while uninformed media will do the lying for you.
Here is a truly amazing story/ video!
A San Francisco TV station has a story titled, "SF company tracking software helps ID suspects" In it, it implies that 3VR was used at the Boston Lord & Taylor location to radically reduce time needed to 'ID suspects'.
We checked with 3VR and they confirmed that their recorders are not used at this location. In fairness, the 3VR person quoted in the article never claims their products were used there. However, the reporter certainly does.
Again, another brilliant job by 3VR's PR team to get this coverage but, my goodness, how misleading is this to readers/viewers who come away thinking that this product was actually used in this case.
A good article on Officer.com about using FLIR cameras to identify the 2nd Boston bombing suspect. The camera used was identified as a Star SAFIRE III.
04/23/13 02:31pm
The Future Isn't Now Once The Today Show says it, it's true
04/22/13 06:08pm
How can the VP of LP for the NRF spout that BS?
Of all the science fiction quotes I've read over the past week (and, of course, beyond) that one is the most troubling. Here is a guy who should be intimately aware of the limitations of camera 'resolution enhancement'.
Unbelievable misinformation.
That's more of the 2D to 3D conversion claims, which would be great if people had perfectly symmetrical faces. Also, that's just for the angle problem, not low resolution.
Here's another article on CNBC from Jim Albers, senior vice president of MorphoTrust. In it, he claims:
"The technology has come a long way in just the last five years and now there are features that will allow users to clean up even low-resolution photos.
For example, if law enforcement only had the profile of an individual in a picture, they could use MorphoTrust's FR software to create an image of what the individual would look like if his whole face was shown in the picture. The software does this by basically copying the visible side and recreating it on the other side. It won't be an exact match, but it will be close to what the individual actually looks like, Albers said."
The WSJ with a cringeworthy column that shows even the biggest name publications can get their 'information' from little more than rehashed PR.
He calls video analytics 'new technology', which is amazing since it's been commercially offered for at least a decade and widely derided for the last 5 years.
And the top providers?
"The technology, from companies with names like IPVideo Corp. and ObjectVideo, is still new."
Even OV's lawsuits are no longer 'new'! And how the hell is IPvideo Corp being cited as a leading video analytics provider? The only connection is that these two companies have done a good job inserting themselves into the recent media coverage.
Alas, this is how the public is being 'informed'.
It was an interesting week on the radio talk circuit here, with a lot of calls for the re-implementation of the thousands of surveillance cameras that were installed for the Olympics, then removed when it was all done. Naturally, there were the typical detractors as well, ranging from the "they're not a deterrent anyway" types to the "Big Brother" and "police state" crowd. In the end, no extra security was added, and the Sun Run went off without a single hitch, with 48,196 participants.
I expect there's a lot of push in most centers to start putting up more cameras - even aside from the marketing claims we're seeing, people still realize that surveillance DID play SOME role in tracking down the bombers, and if nothing else, made it a lot easier for Joe Average to see how the investigation was progressing. It will be interesting, going forward, to see what kind of longevity the calls for more cameras have - whether they simply peter out in some areas, or are actually realized in others after prolonged public pressure... or shot down by public pressure, for that matter.
How many people have lost their mind?
An executive at the NRF claimed that, essentially, Hollywood science fiction enhancement has become a reality:
“Now the ability to enhance the video, and as we have all seen by first images of the Boston bombing, those images were better and more clearer as the hours passed. So the first images of individual faces, that gets enhanced over a computer system,” he said. “Those fuzzy black and white images of people for which we cannot make an identification now have been enhanced to the point they keep refining the image so it’s so clear that they can make an identification.”
I am dumbfounded. I have not heard anyone else make this claim. The closest thing I can think is that he is confusing it with other images being found from other cameras that have a narrower FoV, better angle or higher resolution.
04/21/13 04:18pm
Yet another BRS reference - Minneapolis Star Tribune
"It then alerts officials when something strays from normalcy — people who are loitering instead of getting about their business, people who are jumping turnstiles, and folks who drop a package and then walk away."
In the marathon scenario, there is no 'normalcy', everyone is loitering and there are backpacks and sport bags everywhere. Definitely would've caught any turnstile jumpers though.
Their marketing machine is well-oiled.... that is quite a few major metro papers (and leaders) chanting their marketing talking points in their editorial/opinion pages.
04/21/13 04:05pm
"But Say with a Cray Computer with billions of processes ( Algorithyms) going on at a time, you could run processes that would allow to account for these scenios and make high level dicisions to differ between True & False scenieros."
I highly doubt that.
It was fascinating following a thread on the whole thing on one forum, where guys were constantly posting info/reports/speculation from various news sources they were following... many conflicting. I ended up just following @Boston_Police on twitter as being the only accurate source of information.
You know, the thing that gets me about all the analytics claims, about being able to spot a dropped backpack etc. besides the fact that the sample videos like above occur in nearly-empty areas... is that there's no mention of how the system would discriminate between a "suspicious backpack", and say, a non-suspicious diaper bag... or someone's forgotten shopping... or any number of hundreds if not thousands of perfectly innocuous backpacks one might see with that many people out and about. Not to mention, you're near the finish line of the race where there are bound to be LOTS of gym bags with changes of clothes in them. Can you imagine all the false alarms going off and the bomb squad detonating someone's sweaty runners?
Of course, no sooner will they have everything programmed to watch for suspicious backpacks, than the next bombers will use diaper bags... or shopping bags... or a baby stroller... or any of a thousand other everyday objects that one wouldn't be surprised to see sitting around behind the scenes.
On a side note... the Vancouver Sun Run, which annually draws 50,000-60,000 runners (twice the size of the Boston Marathon, although the Sun Run is only a 10km race), happens today. There was a lot of discussion last Sunday about whether the race would go ahead, but the organizers were adamant that aside from some extra vigilance, there were no plans to alter or cancel the run. And for the rest of the week... daily registrations doubled over last year, with people signing up to show support for Boston, to the point where they're saying they'll have a record number of runners this year.
Boston's police commissioner confirms that facial recognition failed to identify the boston bombing suspects. It's a good thing PR people don't let facts get in the way of their promotions!
04/21/13 04:06am
Bloomberg rhetoric mirrors Globe editorial - almost verbatim:
"..the network now has the ability to 'alert police to abnormalities it detects on the street', such as an abandoned package that is left on a corner."
BRS Labs marketing & PR people do cartwheels every time they read someone singing the lyrics they wrote... :)
^^^ good point, I changed my language! :)
04/20/13 10:43pm
Track down? I thought the guy who owned the boat went out for a smoke and saw blood smears and a tear in his boat cover and he looked inside and saw something. Thermal imaging confirmed his presence maybe..... :)
So it wasn't cell phone triangulation, tracking dogs, facial recognition or any other modern technology that caught suspect #2..... it was smoking.
Chopper born thermal cameras were used to help track down Dzhokar Tsarvaev hiding in boat:
Here's a panomera of the scene from an iPhone 90 minutes before the bombing. It's incredibly detailed, and might be 28MPs.
Wired has a story on how this was used in the investigation.
04/20/13 04:31pm
Completely agree with Brian (both of them).
While it sure seems that the analytics PR folks are working overtime after the Boston Marathon bombing, do a simple google news search for the word 'surveillance'. EVERY story regarding municipal systems since April 16th references the Boston Marathon bombing as the reasoning behind the need for more cameras....
The flawed logic driving this wave of disinformation may boost short-term sales in our industry, but as Mr. Rhodes points out, "fantanstic claims will collateralize everyone in the business when they do not stop anything the next time this happens."
The inflated claims by these attention-whoring video analytics vendors are bad for the entire industry. Their inability to deliver on the fantanstic claims will collateralize everyone in the business when they do not stop anything the next time this happens. (God forbid.)
04/20/13 03:34pm
"might have flagged the Marathon terrorists"
Yeah, it "might" have, just like it also might have suggested a good wine pairing for a pleasant souffle or spontaneously created a bridal registry list.
Good find, Marty. This shows that manufacturer 'investment' in 'emergency PR' has benefits (to themselves at least).
04/20/13 03:29pm
Boston Globe editorial today - spinning yarns:
"Given the horrors at the Marathon finish, Boston officials should be keeping close tabs on new surveillance software that puts emphasis on prevention as well as detection. New systems have been designed to alert officials of unusual behavior. Such a program, for example, might have flagged the Marathon terrorists when they set down their packs and walked away."
Ugh, evidently ObjectVideo can prevent future bombings. Featured on FoxNews (at the 5:30 mark of this video), their Chairman says, "Technology is available today where cameras can automatically detect someone leaving a backpack behind" while showing simplistic scenes totally unlike the Boston bombing, e.g.,
Just terrible...
Tragically, the best evidence is still likely to come from first hand sources:
"Minutes before the bombs blew up, Jeff Bauman looked into the face of one of the suspects, his brother Chris Bauman told Bloomberg News. Two and half minutes after the man dropped a bag at his feet, the bag exploded, tearing Jeff Bauman’s legs apart.
“He woke up under so much drugs, asked for a paper and pen and wrote, ‘bag, saw the guy, looked right at me,’” Chris Bauman said yesterday in an interview. While still in intensive care, Jeff Bauman gave the FBI a description of the man he saw, his brother said."
With such a large crowd, having an eyewitness account most likely did far more to narrow the potential leads than any surveillance. While our industry continues to harp on how important surveillance was, let's remember the role this victim played in identifying the bombers.
Here's a question: How would the investigation be impacted if smartphone evidence was not available?
One thing that will be interesting to analyze carefully (Carlton) is what percentage of the video evidence came from surveillance vs from smartphones.
The original photos were clearly from surveillance cameras but they had very bad angles as they were from store surveillance cameras. It appears that the most detailed / straight on pictures were from smartphones, like:
And
And
The Mercury News, in Silicon Valley, has a fairly balanced article on face recognition and the Boston manhunt. Interestingly, they do not make any claims or insinuations that it is being used in Boston. Rather they discuss the future of it and potential issues, noting:
"The products promise to one day revolutionize crime fighting. Although experts say police won't fully embrace the software until it becomes more sophisticated, they also say that day is not far away."
In fairness, the same exact claims were made in 2005. I remember...
A UK newspaper cites a biometrics manufacturer making a particularly good point about the impact of the brother's youth:
"People's faces change a lot," Mr Carr-Archer said, especially between early adolescence and the mid-twenties. If one of the suspects had applied for a driving licence aged 15 (as is legal in the US), his face may have changed considerably from the image portrayed on it, he said. "If you have a ten-year-old photo on a passport, you can look very different."
Oh brother. Here's a face recognition vendor claiming in a Boston Bombing article, "“Facial recognition is as good as a fingerprint.... Technology itself has improved significantly so that the false acceptance is less than 0.1 percent in an ideal situation.”
If only the world was an 'ideal situation'...
Reuters has a contrarian take, claiming that, "Search for Boston bombers likely relied on eyes, not software,"
Money quote: "They probably just applied a lot of eyeballs to this problem," says Todd Waits, digital investigation and intelligence director at Carnegie Mellon University's Software Engineering Institute and a specialist in forensic video techniques.
Technorati claimed, "Facial recognition technology software like that made by the company 3VR has been useful in helping police and the FBI track down the two suspects in the Boston Marathon bombing."
They offer no evidence how it was used, or if 3VR was used. Amazing, though, again, what a job by 3VR to get the press to give them credit. But holy...
04/20/13 02:01am
[VerifIR] looks cool, but it looks like you need relatively tight shots to get much value out of it. It would seem nearly impossible to effectively scan dense crowds dispersed over a large area.
Paraphrasing a role model of mine, nowhere did he we use the word 'actual'...
Hey he said it can be, not that it should be or that it'll work if you try!
Newest Discussions
| Discussion | Posts | Latest |
|---|---|---|
|
Started by
Donald Maye
|
2
|
less than a minute by Undisclosed Manufacturer #1 |
|
Started by
Scott Zuniga
|
27
|
less than a minute by Undisclosed Manufacturer #1 |
|
Started by
Ross Vander Klok
|
4
|
less than a minute by Ross Vander Klok |
|
Started by
Jermaine Wilson
|
4
|
less than a minute by Undisclosed Integrator #2 |
|
Started by
John Honovich
|
6
|
less than a minute by Undisclosed Manufacturer #3 |
