I have recently seen the virtual mirror concept on you, I tried to take it out and researched it. It seems that the creators have used augmented reality so that people can see the output on their screen. On research I came to know that we identify a pattern on which a 3D image is downward.
Question 1: How are they capable of applying and tracking the person's face without identifying any pattern?
I have also tried to investigate various libraries which I can use to use to show a program. I think many people are using Android phones and iPhones and make apps that use augmented reality.
Question 2: Is there a way that I can use C ++ and try to enhance the program is reality?
Oh, and most importantly, the link to the application is given below:
Try it out: It's a good experience: D
I'm not actually trying a live demo, but the linked video shows that they either Use some simplified pattern recognition (get the outline of the person), or they simply track you based on the initial image (your position / make .
After the video, it is easy to see that there is no real / advanced ARD behind it. Images are simply overlayed or hidden (e.g. In case it is looking at the side due to the track of an ear) and they have not changed (any perspective or again they definitely track the head (or features like ear, neck, etc.) Based on your background and surroundings, this is actually a trivial task.
Question 2: Sure! Many toolets out there premade, but you have to do math as well Like some general image processing libraries can use. Augmented reality usually uses some kind of pattern (for example, a card or page with a known pattern) to determine the correct position and change for the content to be added in the image. To determine the depth / position, there are ways to use device orientation and change in perspective in camera pictures.
Comments
Post a Comment