After trying Wanna Kicks yesterday and got really excited about it. Today I wanted to try it again and spend more time with it to see how good the foot tracking is. From what I’ve read, the tracking algorithm is based on deep learning neural networks and advanced geometric algorithms that allow the app to detect people’s fit and accurately position the virtual shoes onto their feet in a very natural looking way.
The exact algorithm used wasn’t revealed, but I believe it was used by some of a framework that can train deep learning models like Tensorflow.
What is TensorFlow? “TensorFlow is an open-source software library for dataflow programming across a range of tasks. It is a symbolic math library and is also used for machine learning applications such as neural networks.”.
I personally try to make my first steps with TensorFlow and will start educating myself about this topic very soon one TensorFlow version 2.0 is released and it has more tightly integrated with the Keras. Keras is a high-level API to build and train deep learning models. It’s used for fast prototyping, advanced research, and production, with three key advantages.
I have an interest in this topic for quite some time and already started reading about it. This is why I was very excited when I’ve heard about this application, twitted by Tom Enrich a few days ago. I downloaded it straight away and wanted to see how well it works.
Previously I tested out an app called AR Shoe, but it lacked tracking and it’s kind of ruin the experience for me. I really wanted to see the shoe moving with me as I walk and then Wana Kicks was released.
This app is developed by a company called Wannaby, which is specialized in eCommerce and Retail technological solutions that help companies increase conversions, reduce return rates, help visualize products in real life (like used in this app), improve brand image, awareness and recognition, increase customer engagement and disrupt shopping experiences. Goals that any eCommerce or Retail company would like to have or improve, right?
Wannaby uses computer vision that allows its app to track human body parts and deliver more compelling and useful Augmented Reality experiences. Their solutions run neural networks in real time on mobile devices, and as written on their official website: “Leveraging hand-crafted unique data sets”.
This AR Try-On app called Wanna Kicks is definitely a unique and interesting project and it got the attention of the media and my attention as well. The goal is to use advanced technologies to deliver next-generation AR shopping experiences that help increase conversions and purchases and reduce product returns and this, of course, can help companies sell more and grow.
When you try the app yourself you can see for yourself how amazing it feels when you put virtual shoes on your feet. Just imagine that used with more personalization options and the app can show you the product that it thinks you like and then you can browse them and try it out. This is the perfect app for Amazon if you ask me, and I wonder if Amazon or other big eCommerce companies will consider using this tech in their own online shops – I bet they would.
Wannaby also developed an app called Wanna Nails. quite similar to Wanna Kicks, just with the word “Kicks” instead of “Nails”. This app is a virtual try-on nail polish, that allows users to see how different nail polishes look on their real nails in real time. I tried this app and it was awe-inspiring, to say the least.
There is also a short video that shows a Wanna Jewelry app that allows people to try on jewelry as well, how amazing is that. It’s all built on the same concept using deep learning and computer vision and cooperating with companies to help them grow their online businesses.
You can visit their website at wanna.by and check out their project and learn more about the company.
I am overwhelmy excited when I see these type of apps. If I was to develop an augmented reality app, I know that I would prioritize getting into the field of computer vision and focus on Deep Learning (DL) in order to be able to develop applications using advanced computer vision techniques that can be very beneficial for users like the ones that Wannaby is developing.