Entrupy: The AI device that can detect counterfeit handbags | Science & Tech

Entrupy: The AI device that can detect counterfeit handbags | Science & Tech

“I’m going to show you a fake bag so you can see how great the imitation is.” Lucia de la Calle, octagonal at Del Páramo Vintage boutique for second-hand luxury handbags in Valladolid, northwest Spain. She showed EL PAÍS the Louis Vuitton Twist MM bag, a classic that the brand essentially wants to make, and bet it will be reissed year after year in different colours.

Its cost is 3500 euros. “I think my eyes are perfectly trained, but often, even I’m unable to perceive the difference,” says de la Calle.

The woman who bought this bag for 2,500 euros at the flea market brought it to the store with the brand’s signature orange box and original dust cover, as well as a copy of the invoice from its previous owner.

De La Calle has been tasked with doing an initial assessment she calls a “sandwich analysis”: examining the outside, the inside, and then the outside again. In this case, the analysis was simpler than usual because there was an original model in the store.

“First things first: I put the bag on the counter and pick up the chain, because there are chains and then there are chains. Texture and hardness and cold and shine…it all tells you something,” she explains. The second thing is weight: A good bag isn’t usually light. At the time Present, the brand’s website has the exact weight of each bag – but just as we refer to these numbers, so do the counterfeiters.Then you have to look at the details: See if the seams are straight or if there are double threads.Luxury items are always ideal – Those with flaws are discarded before they are put on the market. This bag was impeccable.”

Then de la Calle turned her attention inward. Fakes tend to pay more attention to the exterior of the bag and often neglect the interior: linings, stamps, labels, or finishes can tell the difference between an original and a fake.

“We open it up and the first thing we look at is the stamp,” says the evaluator. Stamp “Comes with a silver-tone embossing exactly the same as the original – Comes with a small mirror with the brand’s signature.” The character looks quite original, “but then I see the first details that raise suspicion.”

Since 1980, Louis Vuitton has included a label with an icon. The code consists of two letters and four numbers indicating the place and date of manufacture. This is a guarantee for customers, who can refer to the code to ensure their bag was indeed made in France in 2006.

“This bag included the code, a good sign,” continues De La Calle, “but instead of placing it in such a way that, when I opened the bag, I could read it, I had to flip the bag over to do it. It’s ridiculous, but the counterfeiters ignore these kinds of details.” .

To complete the analysis, De La Calle notes that the seal inside the bag is very close to the seam, while it is a few centimeters lower on the other bag in the store. De La Calle decides that the woman’s bag was a fake.

“When you tell a customer that their bag is fake, they first say ‘that’s impossible,’ and then they get angry,” explains Sheila Guerrero, co-founder of Del Páramo Vintage. “I get it, most people who come to our store have paid at least 500 euros for a bag, and it’s usually much more than that.” Guerrero says that at least one bag in the store turns out to be fake every week: “The most fake bags are usually Chanel and Louis Vuitton — people are craving classic and timeless bags.”

The fake Louis Vuitton Twist was one such case: The woman who brought it in didn’t believe it might be a fake, but rather got the receipt. “The problem with the scam is that it has become more professional: now they sell you bags that include a bill, because there are companies that even visit bill stores,” Guerrero explains. After all, if someone could forge a bag, how could they not forge a piece of paper. In other situations, “there are people who buy an original bag and then 25 fake ones, and they send you the photos and the original invoice…but then they send you the fake bag,” Guerrero explains. In these cases, Del Páramo Vintage has a device that, via artificial intelligence (AI), can detect a fake in record time.

Artificial intelligence against counterfeit

Entrupy: The AI device that can detect counterfeit handbags | Science & Tech

Lucia De La Calle presents the iPod with a compact device with a huge lens. Then the appraiser opens an app called Entrupy and the system starts guiding it through the authentication process: select the brand; the model if you know it; Choose the type of skin. Next, the app requests a series of images: external (bag on the front, bag on the side, external label logo, zip closure), internal (pockets, labels, the phrase “made in”) and finally, microscopic, where the aforementioned lens zooms in The object is up to 260 times and it is able to identify the most subtle differences between colors or fabrics.

In less than 30 seconds in the most obvious cases, this little tool identifies everything that could take a person hours to do, and will convince even the customer at the worst stage of rejection.

Videoth Srinivasan is the co-founder and CEO of Entrupy. Launched in 2016, the company behind the app aims to eliminate fake luxury handbags and sneakers and is already operating in more than 65 countries.

“The way the system works is that from our app we ask our users to choose the model of the bag in front of them and take a series of photos,” explains Srinivasan from New York. “Then, they upload these images to our cloud. What is in that cloud is a set of algorithms for each brand that allows you to authenticate any model that has that branding,” he says.

“We have more than 2,000 metrics that we use in each image to analyze and help the algorithm make a decision. So this group of algorithms at the end gives a report where they say ‘yes’ or ‘no’, or in other words, ‘original’ or ‘fake’.” The algorithm was trained through thousands and thousands of product images. Over time, the more users use the service, the more photos are included. This translates to more data, and with more data, the algorithm is updated and improved.”

Then there is the microscope lens. “When we started, we didn’t want to have an external device, because as a company you think ‘How many people want a device?'” “But then we realized that, with the quality that fakes now have, the level at which good fakes are being made, we couldn’t be satisfied with being able to spot easy fakes, we had to identify hard fakes,” says Srinivasan. To solve this problem, we had to be able to create images at a level where you can see big differences, maybe not just to anyone, but to the trained eye in our algorithm.”

Entrupy is part of La Maison des Startups, the startup accelerator of the LVMH group, to which brands such as Louis Vuitton, Fendi or Christian Dior, among others, belong.

The CEO explains that companies using Entrupy and customers who pass their handbag through an AI filter can expect safe results due to its high reliability first: “We have over 99% accuracy in correctly distinguishing between genuine and counterfeit products,” Srinivasan says.

To counter any gap between 99 and 100%, the algorithm is programmed to give more false negatives (original bags that the app marks as “undefined”) than false positives: “False positives will be when we have fakes, but we take them out,” says Srinivasan. Admittedly, that’s a worst-case scenario for anyone…except for the guy trying to sell you a fake bag, of course.” In “unspecified” cases, the human technical team is brought in to determine whether or not the product is authentic.

Entrupy also provides a guarantee. “Our goal is that there are no counterfeit products in the market, and that people have more safety when buying products,” Srinivasan says. For this reason, Entrupy offers a Certificate of Authenticity, a type of diploma that can already be seen on the pages of manuals for second-hand for sale (“Entrupy Authenticated Chanel Bag”). “As part of this certification, we provide what we call a financial guarantee,” says the CEO, which means, “We are so sure of the reliability of our proprietary technology that, if we make a mistake, we will pay the money the customer paid for the counterfeit.”

Market full of fake products

The co-founder of Al Paramo Vintage explains that since the beginning of the epidemic in 2020, the number of fakes among authentic handbags has not stopped increasing. Vidioth Srinivasan agrees: “Yes, there has been an increase in counterfeiting in the wake of the pandemic, and I don’t know the official reason behind that, but I would dare say the reason is that, globally, there are more people looking for liquidity, so there are more People are dumping things and more transactions are generally happening. Moreover, there has been an increase in online trading where it is easy to infiltrate fake products.”

According to the Organization for Economic Co-operation and Development (OECD), sports shoes, along with fashion and luxury goods, are among the three most counterfeited products. According to data for 2019, the international trade in counterfeit and pirated products was estimated at $464 billion, or 2.5% of world trade. Apps and pages that sell used products between individuals and social networks, such as Instagram, are the places where the supply and demand for these items has increased the most, and where the buyer has the least guarantees and security. While classic and timeless luxury brands, as Sheila Guerrero rightly points out, are among the most counterfeited, criminals are also increasingly focusing on limited editions, the latest models, or smaller brands.

“Luxury brands often have a strategy to limit certain products,” Guerrero explains. “This often leads to a spike in the flea market. In the end, second-hand prices are the same as in retail, in both jewelry and handbags, and there are usually no bargains. My only advice to those who want to buy a used product is the following : Beware if someone offers you Chanel for 500 euros – most likely it is a fake. ”

Source link

Leave a Comment