Scientists Create Easier Way To Embed Objects Into Video 236
Ashutosh Saxena writes "Stanford artificial intelligence researchers have developed software that makes it easy to reach inside an existing video and place a photo on the wall so realistically that it looks like it was there from the beginning. The photo is not pasted on top of the existing video, but embedded in it. It works for videos as well — you can play a video on a wall inside your video. The technology can cheaply do some of the tricks normally performed by expensive commercial editing systems. The researchers suggest that anyone with a video camera might earn some spending money by agreeing to have unobtrusive corporate logos placed inside their videos before they are posted online."
Re:The death of advertising (Score:5, Interesting)
I was afraid this might happen (Score:3, Interesting)
When American football television broadcasts started featuring real-time "underlays" of such play-by-play landmarks as line of scrimmage and first down mark, a worried little voice at the back of my head wondered if someone would use this technology to underlay advertising. I think I've seen just such things (i.e., digitally-projected advertising hoardings in the video background, even logos "projected" into the playing field). Now this kind of stuff will be easy and ubiquitous.
As little as we can trust digital visual media now, it'll be even less trustworthy.
Re:Stanford sold out (Score:4, Interesting)
I dunno, it seems that this knowledge is directly applicable to all kinds of serious real-world problems involving computer vision, particularly automated car driving (the Stanford [wikipedia.org] connection might just be a coincidence, but there's a lot of overlap).
Nothing new here (Score:3, Interesting)
The technology to do this was commonly available in the mid-90's.
Re:It will, and does (Score:3, Interesting)
It's the exact same principle. The only difference is the magnitude of the effect.
Re:It will, and does (Score:5, Interesting)
In the store people are then more likely to subconsciously reach for Tide or Tylenol (despite the fact that generics are composed of essentially the same active ingredients) because they are familiar.
I guess once again I'm outside the bell curve, as I use generics almost exclusively. Tylenol? Doesn't work. Advil? Hell no, generic Naproxin Sodium is 1/3 the price. Tide? Yeah, becasue I haven't found anything that works as good. Listerine? Yes, that brand was shown to reduce incidence of gum disease which I suffer from, and the generics are watered down, you can tell because they don't burn as bad (yes, I did try them and found them wanting).
However, NEGATIVE ads work well on me. Sony's rootkit bit me when my daughter trusted BMG and I'll never buy another Sony product again. I spent so much time under the hood of my Mustang in 1970 that I never again bought another Ford. Tyson Foods burned two dozen Mexicans alive in Georgia in the 1980s because they chained the fire exits shut to keep them from stealing chicken parts (a manager spent 2 years in prison for twenty five horrible deaths) and I'll pay MORE for generic meat than buy Tyson.
And some ads are so annoying that I deliberatly avoid the products.
You would think that the corporates would learn. It's an old adage that if you're happy with a product you MIGHT tell a friend, but if you feel like you've been ripped off you'll tell everybody.
Re:I was afraid this might happen (Score:4, Interesting)
They already do - those advertising boards on the side? They're electronically generated - have been for years. That's why when you see something played in another country all the adverts are in english.
Re:The death of advertising (Score:3, Interesting)
So, legitimately, how powerful is a wall-hanging logo for Pepsi in some random goofy youtube video ACTUALLY going to be?
A lot more than you realize.
Advertisement is one of the most heavily researched areas of our lives. A good fraction of psychology research is, directly or indirectly, related to the effects and effectiveness of advertisement. While it is almost impossible to correctly estimate any specific ad or campaign, the general effects of advertisement are extremely well researched.
So you don't consciously register the ads anymore. Do you think the advertisers care? No, not in the least. They were never targeting your consciousness anyways. Advertisement is about embedding stuff into your subconscious mind - desires for a specific product, good feelings about a specific brand, that kind of stuff.
Re:oh look, its the anti-ad crowd (Score:3, Interesting)
I am a member of a number of websites with no advertising, even large ones.
I even pay subscription fees to some.
The fact that there is so much advertising that I tune it out has little to no relevance to your desire for me not to share my opinion, or your belief that my opinion is going to shut down Slashdot... or whatever.
Frankly, I made my income for 5 years off advertising, but we did it in a way that was not stupid banner ads. There were active discussion forums about products related to our content, where people got together to purchase the product for a volume discount and in exchange for organizing it, the site took a moderate cut.
The people got a better deal than they would have without that service and the site was able to make a decent income doing it.
Of course, that was ages ago, but the fact still stands that there are unobtrusive means to conduct business.
On another hand, I block advertising as a side effect.
I block third party cookies, I use NOScript and FlashBlock for legitimate security reasons. The fact that this effectively kills about 95% of advertising isn't really my problem.
They want to drop cookies on my machine and make flash animations fly all over and I don't allow unauthorized flash animations.
Google ads still appear for me, and those are often quite relevant, but one issue is that I don't purchase things online through random retailers FOR SECURITY REASONS, so those ads have little to no benefit to me.
If I want to buy something online, I'll go find a retailer rating website where I can choose a few good retailers to check out and make an educated decision from there and only purchase from reputable stores, regardless of the advertising.
I never called anyone stupid or called for the decline of slashdot, but I won't pretend that ads are super-duper effective, just so that you don't get upset.
Re:Help Please (Score:3, Interesting)
Re:Nothing new here (Score:1, Interesting)
It's a different approach (Score:1, Interesting)
I looked to me like the point of the project was to allow any user to simply draw an area on some plane in the scene, and have the computer automatically track it through the video and matte out objects in front of it as well as estimate lighting changes. This can be any area, and doesn't have to be of a particular solid color. My guess is that you can easily underlay things on the football field because most of it is well... green.
Think like a corporation (Score:3, Interesting)
The researchers suggest that anyone with a video camera might earn some spending money by agreeing to have unobtrusive corporate logos placed inside their videos before they are posted online.
More like Google will insert said logos into said videos upon your posting them on YouTube.
Re:Yeah, that'll work (Score:3, Interesting)
I wonder if it would be just as easy to take them out?
Re:It will, and does (Score:3, Interesting)
I particularly liked how a CSI:NY episode matched the marks on a dismembered body's bones to the blade of a particular brand of cordless reciprocating saw, then the same saw was advertised in the commercial break, just in case you had some bodies you needed to dismember.
Re:The death of advertising (Score:3, Interesting)
The point is that the ad placements do tend to get me to think about the brand.
Re:The death of advertising (Score:1, Interesting)
But you're smarter than the average bear doing that ;) Most people will equate a brand they haven't heard of, or heard much of, with bad. And generic ANYTHING has to be bad.
Generics (Score:3, Interesting)
(despite the fact that generics are composed of essentially the same active ingredients)
Have you bought generic or third party hardware before? I'm thinking of a replacement car fender a body shop once tried to sell me. The steel was thinner than the original steel, and much more flexible. The rolled form of the fender didn't exactly line up with the rolled form of the original. The factory fender, on the other hand, was a perfect match to the original part. I'm sure the car would have *looked* like it was supposed to, but if you examine the fit and finish up close, it's evident that it's not a perfect match.
I think of that every time I buy a generic product. You pay less for a generic product for many reasons: the lack of a patent license or advertising are only two attributes of the lower cost. The materials may be inferior. The measurements may not be as precise. How many of those "low quality" attributes carry over to generic medicines? Am I getting the correct dosage? Is this product cut with food-grade corn starch or with clay dug out from the field next to the factory? Did they copy the binding agents that keep it from irritating my stomach, or did they use their own binding agents that might not work for me, and do those new agents have any interactions with the medicine?
I'm not saying generics are bad or ineffective -- I usually buy them. But I always think of the quality differences on the products I can see, and wonder about those differences that I can't.
Re:It will, and does (Score:3, Interesting)