TL;DR
This is a proof of concept for the technology we implement behind the scene. Our software and solutions use various technologies to solve business problems. One of the most interesting tech now is to enable machines to learn. Scroll below for machine learning demonstration.
A simple example, given 9,999 images of cats for a machine to learn makes it able to say “Cat” when human upload a different cat image.
This kind of learning can be very useful for business. For example, we could input thousands of sales data for prediction. Or, automatically find patterns on which aisle is crowded at which time of day by which type of customers. Or, inventory knowledge feed from system based on various supply chain data reading. And many others.
We prove it for you
Let’s take a look how machine learning works.
As any kind of learning, machine learns in 2 steps. First, we train the machine (training). Second, the machine perform task we give based on what it learns (performing).
We have prepared a machine for you on the demonstration below. This machine has ability to learn and perform what is it trained for. You can try it. Give the machine enough sample data (e.g. images of hand gestures) so it can recognize certain pattern (e.g. palm, fist, ‘metal’ sign, one, two, three, etc). We will make the machine does different things for each pattern.
Demonstration
On this demo, we provide 3 patterns. You may give each pattern the sample data it needs. To ease the process of data collection, we have prepared a button which you can press and hold. Place your hand on your webcam, make a gesture, and photos will be repeatedly taken as long as you hold the button. Once you think the data is enough (maybe around 30-50 frames), release the button and continue to make different gesture of the 2 other patterns. When you are done, congratulation! You have just trained your machine!
Now, to see the machine performs for you, show a hand gesture to your webcam and let the machine determine which pattern it detects. Then, you can see it behaves as it was trained. Enough talking, let’s see it in action.
