Need autonomous driving training data? ›

Conversations in Machine Learning: An App That Sees, Searches, & Shops

Conversations in Machine Learning: An App That Sees, Searches, & Shops

This is another installment of Mighty AI’s “Conversations in Machine Learning” blog series. Each week, our content human, Cassie, shares a summary of a recent conversation we had with a machine learning team and potential customer—what they’re building, how they’re handling training data today, etc. Read more about the series here.

Hello, fangirling again over here. This week I’m featuring a technology that, when I learned about it I said, man I love when people think of stuff like this! Great idea, you technologically inclined people you. Love those brains of yours.

So imagine you’re flipping through a magazine and see an ad for these great shoes and you’re like, those are perfect shoes. I want those. Get on my feet. You could head to Google, search for the brand, pull up their website, browse for the shoes there… Or you could just snap a picture of them with your phone and let the machine learning-based app we’re highlighting today do all the work for you. It would take you to a custom digital experience in which you can learn more about the shoes and purchase them if you wish, minus the multiple taps and queries and screens of the old-school option. How’s that sound, reader? Sounds good, I know.

This visual search app is powered by image (and video) recognition and deep learning technology, and it works for ads you see in print, but also on TV or a billboard. And as visual fashion search is a major focus of the company and app, you can also take a picture of a real-life outfit and get recommendations on where to buy the clothing (or similar styles) in the photo. YES.

This is a sweet deal for consumers, obviously, but it’s great for brands and businesses, too. By serving targeted campaigns and entertaining experiences for folks expressing interest in their products in real-time, companies can engage customers in a fun, non-intrusive way. Win-win. Brands don’t have to specifically work through this app, either—they can integrate the technology into their own existing apps. Nice!

So let’s talk training data, because that’s what we at Mighty AI care a lot about and that’s why this cool company and their Computer Vision Engineer and Principal Image Processing Engineer are talking to us (and why I have this insider info). To train their visual fashion search algorithms, their computer vision team is currently annotating data internally. And as we’ve covered many times and as I know you know, that ain’t scalable. And it takes valuable team members off far more critical projects. They need a training data solution stat.

What Mighty AI and our excellent tasking community would do for them is two-fold: 1) categorize fashion-related images into 1 of 10 total categories, and 2) create bounding boxes around fashion-related items within images, and tag those items with identifying terms. So sorta three-fold I guess, but whatever.

It kind of stresses me out to think of them trying to do all of this themselves on top of building and refining this amazing technology, so if you want my advice, anonymous engineers, switch to Mighty AI’s TDaaS solution. We’ll do all your training data stuff and save you boatloads of time and many headaches.

K that’s all but talk to you next week! Follow us on Twitter and LinkedIn in the meantime.

image credit: London Scout via CC0 1.0

Note: Prior to January 10, 2017, Mighty AI was known as Spare5. While Spare5 remains the name of our consumer brand and application, we’ve relaunched our business-customer side as Mighty AI, which also serves as the parent company under which Spare5 now lives. Some posts on spare5.com have been updated with the new company name to ease confusion.