Recogneyes

Inspiration

I wanted to bridge the gap between human's and machine interaction using computer vision as a basis

What it does

It tracks the hand's movement and attaches a point to a finger which is used to draw the image. The scribbled image is then fed into a custom model built on Clarifai's API, which identifies 2D sketches made by humans. The predicted image is then used to search for an image using Google's custom search API, unlike the initial scribbled image.

How I built it

We stand on the shoulder of giants. My team and I merely brought different codes together to form a cohesive project

Challenges I ran into

Some of the opencv is difficult to get adjusted to. This project is my first experience with both Python and OpenCV

Accomplishments that I'm proud of

For the most part this project works well and I am glad that despite being new to the languages,libraries and APIs we were able to create a product.

What's next for Recogneyes

In the future, I would like to optimize the hand detection algorithim combining the effects of background subtraction as well as HSV histograms to create a more distinguished image for the computer to track.

Built With

• clarifai
• google-custom-search
• opencv
• python
• tkinter
• wikipedia-api