Virtual Dummy To Try On Clothes
Roland Piquepaille writes "BBC News reports that Toshiba is working with a Japanese software company to create a 3-D fashion simulator that will allow virtual modelling and coordination of clothes, cosmetics and accessories in real time. This means that by as early as 2006, you will no longer have to contortion yourself in a minuscule fitting room. 'Video cameras snap the shopper, then clothes and accessories are selected and displayed immediately. The process of turning the images of the shopper into photo-realistic avatar -- or virtual representation -- happens in real-time.' This summary contains more details and references. It also contains images of a virtual model trying different clothes and accessories adapted to different backgrounds."
Oh, great. A computer simulation of my big, fat butt. I am overcome with joy at the prospect.
Boobies never hurt anyone. - Sherry Glaser.
This means that by as early as 2006, you will no longer have to contortion yourself in a minuscule fitting room.
...
And there goes the hidden cam live internet feed porn business
"A door is what a dog is perpetually on the wrong side of" - Ogden Nash
People don't try on cloths to see what the cloths look like. They can do that by just looking at them. People try cloths on to see how they fit - ie, how big their boobs/asses look.
Quid festinatio swallonis est aetherfuga inonusti?
Africus aut Europaeus?
Is my coffee defective, or is this concept just underwhelming?
"Derp de derp."
After these roll out, how long would it be until the software is modified to bias how you look?
It could make you more "perfect," and you would buy that dress!
I don't think they should have used "start with a naked model" and "seduce even men" in the same article. :-)
Learning HOW to think is more important than learning WHAT to think.
I defy anybody to be able to make my clothes match up though, what with this being /. and all. We shall not be cool!
Gee, I hate when I have to contortion myself anywhere. To even have to contort myself, grammatically correctly no less, would be even more brutal.
Canadian Cynic, canadian politics is less boring than you
My wife has been calling me a "virtual dummy" for years! Think I should apply for this job?
"Freedom means freedom for everybody" -- Dick Cheney
The problem with this is that fitting rooms are to see how the particular garment actually fits. It's one thing to see how it looks on you, but to figure out whether you need a medium or large, you need to try the clothes on.
I am feeling fat and sassy
How long before large shopping chains will start to hack these programs to alter the shopper's virtual body to fit the clothes better, so they can make better sales?
If programs would be read like poetry, most programmers would be Vogons.
First of all, the same size is never the same size is never the same size. If you really want to know whether the clothes fit, you have to put them on. A second, related point is whether the clothes are comfortable. No matter how good they look, in the end you need to wear them.
The net will not be what we demand, but what we make it. Build it well.
> They need to finish perfecting the ones that take off their clothes FIRST.
First karaoke, now this.
Sheesh, evil *and* a jerk. -- Jade
Can you put it in chains ?
"However beautiful the strategy, you should occasionally look at the results" - Winston Churchill
I take the summary (about to leave work, can't RTFA) to mean that "Toshiba and a Japanese software company" haven't started the project, much less come up with a working prototype. Since getting stores to buy and use (and therefore test) this sort of thing would take a couple years, I'll believe this isn't vaporware when it's in the first store.
Maybe they're working on my flying car, too.
It's nothing but crumpled porno and Ayn Rand.
I don't see how you could buy clothing without trying it on.
How many times have you run into clothes that are either mislabeled, or cut too small?
I've learned the hard way that it's always better to spend the 10 minutes trying stuff on in the store, rather than spend an hour on a return trip.
So I think this modelling scheme is useless. Unless of course that XL shirt is really an XL and not an L.
Also, you need to experience how clothes feel. Do they look cool when you look in the mirror? Do they feel good on you? Does that sweater itch your arms? A model can't tell you this.
I'm all for progress, but it has to be practical. Especially when it involves me spending money on something.
wbs.
You always need to try on clothes.
Huh?
How long before the representations are secretly tweaked (displaywaist = size 6) so that the shopper will "look good" in the clothes? And I can see the tie-ins with advertisers, with avatars saying things like:
"Hi shopper, this is what you look like now, but here's what you would look like (shrinks waist) if you go the XXX diet!".
Just wait until the kids start hacking it!
What is the inverse of the Matrix?
Interesting, though seemingly unremarkable - I believe Pixar already has a program whereby you can scan in a McCall's pattern, and it will sew the garment and fit it to one of their characters...
Philip Sandifer's academic website
The Barbie dress up program.
But who the hell can tell if something looks good on a screen ?
There is a LOT more to looking good than just "the clothes" or the style of the clothes.
Anyone who owns an Armani or HSM will attest to what I am talking about.
Because its really so hard to walk ten feet and try on new clothes. On another note when I try on clothes I like to see how they feel(ie too tight/loose), and I like to see what it looks like up close in a mirror, also what if the computer some how manipulated the clothes to make you look better than you really do look? Your probably just going to end up trying them on eventually anyways. And besides, is it really that difficult to visualize yourself with a shirt on ? I should hope not. This is a stupid idea and it will never work.
If the dollar is an "I owe you nothing", then the Euro is a "Who owes you nothing." - Doug Casey
I fully expect that most retailers would have a version of the software in which your big, fat butt doesn't look quite as big or fat in the clothes you're modeling. People want to buy clothes that make them look good and it is the job of the software to convince them that they look good in those clothes.
Mmmm.. Donuts
Land's End (www.landsend.com) has had a jr. version of this for a long time.
The model shows how dumpy I really look, regardless of color or outfit. As a result of experiencing the preview, I haven't bought anything from them in a couple of years.
So using this technology this company is going to sell more clothes why?
Live Long and Prosper - Thanks Leonard. You are missed.
To see it in action, go to their site. and click on 'My Model' in the upper left corner.
Which is a 16 year old cheerleader.
Cool. Does this mean we will get to see naked photo-realistic avatars of all the other shoppers too??
;-)
If so, I'm going to start taking my girlfriend underwear shopping more often
I'd rather be a conservative nutjob than a liberal with no nuts and no job.
> After these roll out, how long would it be until the software is modified to bias how you look?
"Honey, do these pants make my ass look big?"
"No darling, it's your ass that makes the pants look big."
Sheesh, evil *and* a jerk. -- Jade
There's a company in Burbank called CyberFX which has been doing 3D scanning for years using Cyberware scanners. They did all the obvious things with the technology -- reverse engineering, prototyping, sculpture scaling (they did the massive baseball glove a PacBell park), porno (scanning and sculpting rich guys' girlfriends), scanning actors for CG doubles in movies.
What they really hit it big with, though, is dressmaking dummies. In the past, dummies were built by hand, and they were just not very good. They didn't match people very well, and each one was different. Now, (say) DKNY sends their size 4, 6, 8, 10 models to CyberFX, they get scanned, and perfect copies are sent to all the dressmaking facilities around the world. Actors have dummies made that match them perfectly, so wardrobe departments can make clothes that fit perfectly.
Dick Cavdek, who runs the company, has come up with significant mechanical advances on dummies, too, so that they are sturdy, light, and can be broken down to be shipped easily.
I went by there a few years ago, and was absolutely amazed by how one guy just revolutionized an industry.
thad
I love Mondays. On a Monday, anything is possible.
'Video cameras snap the shopper, then clothes and accessories are selected and displayed immediately.
And I suppose the virtual models tell you which bits chafe?
"Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
Reading the article, I notice that the digital model starts off nude, then has clothes rendered onto it. In order for these photorealistic images of a person in new clothes to be generated, it would seem that the software requires a nude scan of said person. Of course, the person could be scanned while clothed, but you'd really just be rendering new clothes on top of the old ones (which, in the software, would take on the rigidity of flesh). It might be possible to design an algorithm to "strip" the scans, but the accuracy would be limited by the varying bagginess of whatever the person is wearing.
Naturally, privacy and convenience concerns arise when one is asked to submit to a nude, full-body digital scan in order to use the new fitting system. The store could keep scans on file, making this a one-time affair, but unless a secure crypto system (wherein only the customer possesses the key) is implemented, the potential for creepy abuse is enormous.
One solution I can think of, however, is to do the scan in a private booth while the customer is only wearing underwear (most undergarments being tight-fitting enough as to not affect the image of the clothing being worn). This is still more convenient than going through a stack of clothes to try on, as the user needs only "change" once. The model for the system could be deleted after use, or the customer could elect to store it on a USB memory stick they bring with them, updating it only occasionally as their physical appearance changes significantly (it could even be stored centrally if a department store chain, or better yet a consortium of them, decides to implement a truly secure system).
It's true that this system doesn't offer as good a "feel" for clothing as actually trying on outfits. For men or women on the go, however, it could drastically reduce time spent clothes shopping. Imagine browsing through the latest fashions at home, picking out a few you like, then heading to the Department store, where they have everything you picked out, in your size, ready for you to try on (and you'll still want to, if for no other reason than to gauge the comfort of the clothing and verify the program's accuracy). An hours-long shopping trip could be reduced to a managable 10-15 minutes.
Of course, my wardrobe consists mostly of items from Goodwill or Thinkgeek, so this is of little utility to me. Nonetheless, it has some potential to make life a lot more convenient for my girlfriend, my sister, etc.
Anonymous Luddite: "What do you think of the dehumanizing effects of the Internet?"
Andy Grove: "Not Much."
Hey, it's better than looking at a picture in a catalog, and people have been buying clothes that way for years.
- None can love freedom heartily, but good men; the rest love not freedom, but license. -- John Milton
back in 1970 i helped out with a somewhat similar project at the Royal College of Art in London (i was writing 3d mapping software at the time). Some lucky students got to digitise real-live naked girls. .. the hardware wasn't fantastic but it did work..
but it was a great excuse to closely examine some very pretty girls' bodies
the 3d-figures were used in a program that "hung" student's dress designs on them
Paul
www.opencouncil.org
Open
I worked for an hosting service and we hosted JC-Penny's version of tis. The virtual manequin, it was all 3D and everything, so I'm not too sure what's so inovative with this. You could set your manequin to match your shape (required some hard honesty, I'm sure) and put clothes on it to see how they fit. I wonder if they still do it, I know for sure that we aren't hosting that anymore.
Mind the frickin' laser...
During the dotcom years, I remember meeting with a client who represented a company that was going to put 3d scanning stations in malls for this very purpose.
"It's simple!" sez their Marketroid. "You go into the booth, strip, get a full body scan, and then we upload the 3d model to our Microsoft Passport(tm)-like service. Then, any participating online retailer will be able to recommend sizes, show you how you personally would look in any outfit, and do dynamic upselling by showing how much better the Gucci looks."
They even had a plan to implement realtime draping/rendering software so you could get photo-realistic images of yourself in those clothes. They thought that boyfriends would finally be able to buy clothes for their girlfriends. They thought that geeks would start getting color coordination.
My first thought was: how many people are going to let some bizarre company photograph them in their underwear (or less), just so that company could better market to them?
My next thought was: nobody reads the disclaimer they sign. I'll set up booths in malls, and run a voyeur web site, and people will even pay to model!
My next thought was: Jesus, I've gone as insane as these lunatics. I need a drink.
Needless to say, they burned through a lot of money, and it never went anywhere. Some guys got some nice SGI hardware out of it for the software side. The "idea people" probably got nice fat salaries for a while, and then had to go back to selling life insurance or flipping burgers or something.
Jesus, do I miss those days!
Eloi, Eloi, lema sabachtani?
www.fogbound.net
I attended a presentation by a company trying to do something like this in Pittsburgh a few years back when I was hunting for a job. I'm pretty sure that company died soon after, but I can't recall their name...
;)
They were working on a mathematical system to model dropping cloth over a surface, so online stores could let people try clothes on 3d models with their dimensions. I thought it was an interesting idea, but decided that I didn't want to interview with them as I expected they wouldn't make it. I was just hoping the cloth modeling they developed would end up making it to video game development
I don't try on clothes to see how they look - I can see what hideously colours and styles they have just by looking at the rack. I try on clothes to find out how they feel - if it isn't comfortable, I ain't wearing it (even if my bum doesn't look fat...)
In early 2000, I consulted with a company that handles much of the B2B for the fashion industry. This functionality was discussed as part of a B2C add-on. They wanted it, and were trying to price it, but many factors made it difficult:
1. The audience was mostly female. Most men would not bother with the system. And women were less likely to be buying on the web. So the ROI was difficult to justify. (This and some of the following include sexual stereotypes. There is a reason they are sterotypes.)
2. Most women will lie about their body size. Could we automatically adjust the virtual bodies up one size? Yes, but that would upset the honest women. Would women be honest when their purchasing decisions depended on it? Since the system was not built, this was never answered.
3. Would women even enter all the information needed? Height, weight, waist, inseam, bust, shoulders, arm length, neck width, circumference of biceps and thighs. Think of all the measurements that a tailor makes. Now expect women to enter all that for each website that uses the system, and update it when their shape changes. (Very few people are the same size in January after the holiday eating as they are in September after Summer's outdoor activities.)
4. Would women be concerned that there is a complete record of how their body changes? My mentioning this was a little ahead of the times, as privacy concerns were not in your face then. But would you like a system that remembered every time you added a few pounds?
5. The model would need to show how clothes drape over the body form. We would need incredible horsepower to run the system. We already knew all the details of the fabrics as part of the B2B system that helped designers choose appropriate fabrics for their creations. That part was just programming, but 3D modeling is CPU-intensive. (I recommended hiring some game-engine programmers to optimize the system.)
6. How are the clothes shown? Do we offer choices for whether a blouse is tucked in, and how tightly? How many buttons are fastened? The width of a belt, and exactly where it is worn?
7. Could we show several products at the same time? This one had us baffled, especially if we were to combine products from several companies. The company hoped to set up a single website that the branded websites would pass buyers. I do not know if the fashion companies would have done this. The largest companies have a complete line, so would prefer to buy the technology for their own website.
---
The company sold software. I was recommending that the software be free, but that the company take a (very small) cut of each transaction. They were already discovering that people were using their free-but-limited version to not pay for the full-featured version, even if the customers had to type much of the information in the comments. The company asked me to make it impossible to use the free version for the main tasks that were in the full version. I recommended making the full version (their cash cow) free, but providing a central clearinghouse to handle the transactions. My recommendations were presented to the president of the company. The company was bought later that year and I have not heard from them since.
I just looked up the company that bought them, and they have several press releases this month about winning new customers for their "product lifecycle" software, so they are still active as software sellers, but they do not own the B2B fashion market as I recommended.
I spend my life entertaining my brain.
I smell a lawsuit comming on. As soon as SCO realizes these people are using virtual dummies....after all that would be a virtual Darl, right?
. Quit playing Monopoly with Bill. Switch to one of many non-Microsoft products today.
My Virtual Model Inc. has set the standard for online apparel shopping. Two core technologies - the My Virtual Model(TM) Dressing Room and My Virtual Model(TM) Fit - enable consumers to 'try on' clothes on the Internet. 2002 saw the introduction of a third product, My Virtual Model(TM) Imail. This innovative marketing tool offers a unique, fun and simple interactive My Virtual Model(TM) Dressing Room integrated in an e-mail.
For merchants, deploying My Virtual Model(TM) technology improves bottom-line profitability by increasing revenues while cutting costs. Shoppers using My Virtual Model(TM) solutions spend more, buy more and return far fewer items - resulting directly in reduced shipping and handling costs.
My Virtual Model Inc. was co-founded by Louise Guay, Ph.D., and Jean-Francois St-Arnaud. In 1990, the two entrepreneurs joined forces to build one of Canada's leading multimedia agencies, Public Technologies Multimedia Inc. (PTM). The first release of My Virtual Model(TM) technology in 1997 was a major breakthrough in online customer service. In 2000, PTM became My Virtual Model Inc., in recognition of the company's focus on the online apparel shopping market.
--- Worst tagline ever.