Slashdot Mirror
Surveillance Cameras Used To Study Customer Behavior
An anonymous reader writes "Technology Review reports on a startup with software used by stores to track, count and log people captured by security cameras. Prism Skylab's technology can produce heatmaps showing where people went and produce other statistics that the company claims offer tracking and analytics like those used online for the real world. One use case is for businesses to correlate online promotions and deals — such as Groupon offers — with real world footfall and in-store behavior."
its not IF but WHEN.
everyone who has an interest in 'tracking' will want to be able to ID people and know where they are.
govs want this, businesses want this, 'law' enforcement wants this.
the only people who don't are the people; and they have no power anymore in the western (or eastern!) countries.
its been said each generation, but its true here: I fear or the world our kids are going to inherit. it does not sound at all like a world I want want. I can see where things are going. Do Not Want.
--
"It is now safe to switch off your computer."
Target has been doing this for many years; in house. They have had software for years which spots people who move around like a shop lifter and many years before that they kept logs of the parking lot car plates so they could ID a crook from anywhere in the store by following them back to their car. This was not widely known either... now people probably assume the parking lot has cameras but back in the 90s not so much. BTW, they are interested in ID of people by their walking gate and I would be surprised if they weren't supporting such research along with the UK.
Target also has one of the best computer forensics teams in the nation, way better than the FBI and they even do work for the government. All in house; if they didn't contract it out we'd probably not know about it.
I guess I better start behaving oddly all the time whenever I'm in a major city. Maybe that's why there are so many oddly-behaving people in major cities ...
I think this story illustrates that the RFID is completely unnecessary.
It's better to vote for what you want and not get it than to vote for what you don't want and get it.
- E. Debs
You know with all the outrage Slashdot has with this type of stuff...
You got to admit it is pretty cool use of computer science....
If something is so important that you feel the need to post it on the internet... It probably isn't that important.
Football? What?, oh wait, misread. Someone tell me I'm not alone in that error.
At the end of the day does this yield better results than counting sales at the close of business?
More complicated results, perhaps, but after analyzing traffic patterns all day long studying dwell time at displays, does it really yield anything useful that the store owner can actually act upon, re-arranging the displays, etc?
And if they do act on the data, it will almost certainly be to benefit one product area vs another. Will there be any net gain for the store as a whole?
Won't wholesalers with clout demand the data and push hard for the best locations or shelving decisions? If you have data, you are going to be forced to share it sooner or later, and when everyone is rushing past the Laptop counters to get to the TV display area, is there anything short of re-arranging the store you can do about it? Won't Dell, HP, and Apple, insist on being on the high traffic routes? Didn't the store owner just lose control?
And at the end of the day, is it different in any way from just tallying sales ?
Sig Battery depleted. Reverting to safe mode.
The Hentai row in the comic book store isn't as private as you think it is.
Have gnu, will travel.
Doubtless there will be volumous FUD in relation to this technology, however I don't see there being a problem here. Consider a book shop. This technology could be utilised to provide the book shop with verifiable information regarding what the most popular categories of books are and thus enable them to make informed decisions about which departments / shelves / sections to expand and which they could safely contract. The end effect being that customers gain access to a greater variety of books concerning their favourite subject and the store is empowered to make the most efficient use of the space available to them. If it were possible to track and more significantly identify people via this technology then I'd agree that there would be privacy issues that ought to be dealt with prior to the system being used, however the linked article quite specifically mentions that scenario as being deliberately impossible.
So its time for a makeover.
Have gnu, will travel.
step 1: observe correlation: the more time people spend in your store, the more they buy
step 2: optimize placement of stuff so that people stay longer in store
step 3: profit!
side effect: waste time of peoples' time.
Hey don't blame me, IANAB
If you *steel* a food, you have bigger issues.
---- Booth was a patriot ----
As a marketer I can tell you that we have been doing this in one way or the other for the last 50 years or so ... the only thing interesting here is that they're doing it with more advanced algorithms. This is stuff we had in the basic course of consumer behavior. The only real difference is that today you OFTEN do these studies in person instead of having computers to do it for you. Results and findings will most probably be the same from this as it is from regular observation studies.
Besides, when doing observation studies the point is not to disturb consumers so they are usually done from a surveillance room or such to monitor consumers movements. Marketers are very seldom interested in individual interactions but when consumers do stand out from the norm it's good to have the possibility to interact with said consumer.
I see this as a possible solution for very large shopping centers and their likes but I don't think there's anything special about this thing in particular.
Stay home. Buy online.
Oliver's law of assumed responsibility: If you're seen fixing it, you will be blamed for breaking it.
I think this story illustrates that the RFID is completely unnecessary.
True story: a local tv news station was busy showing off their latest scare-piece on RFID technology some months ago. The anchor phonetically pronounced it "ar-fid". *head-shake*
I only post comments when someone on the internet is wrong.
Here. And the data is already in numerical form.
You're complaining that he pronounced the acronym? That's a pretty common thing in English, at least in the last few decades.
Unless you say L-A-S-E-R, N-A-T-O, A-I-D-S and S-C-U-B-A, among others--and though I don't know you, let me say I officially doubt it--then you're really just judging somebody over his decision on where to draw the line between acronyms that should be pronounced and those that should be spelled out. Frankly I have no problem with pronouncing any acronym that pronounces smoothly. (Yes, Slashdotters, I typically pronounce "SQL" -- run in abject horror!)
I don't live in the U.S., so I wouldn't know where to send people (e.g. bricks and mortar stores) to find this particular book. It isn't a huge bestseller, and thus probably not in stock everywhere. The Amazon link has a lot of info + customer reviews of the book, so I posted that. Feel free to buy (or not buy) this book wherever it is you shop for books.
Why did the chicken cross the road? Because Elon Musk put an AI chip in its head.
Check out the difference between an acronym and an initialism. Do you also say "YOO-suh" (USA) or "SEE-oh" (CEO)? The examples you gave are easily (and obviously) pronounced, and most style guides will say that you can spell them with lower case letters.
I've got an aunt who used to work in sales for Oracle. She also pronounced "SQL", which baffled me, because I thought the "sequel" pronunciation was reserved for the Microsoft product. And yes, I shall now run in abject horror.
I only post comments when someone on the internet is wrong.
Like a movie, sequel.