Shark Bytes

Screenshot 2
Screenshot 1

Inspiration

I was inspired to use the theme of the hackathon, being about sharks, and create something quirky. So I took this opportunity to make a fun project which makes people aware about sharks.

What it does

This project helps you explore the wondrous underwater world of sharks, by showing fun facts about sharks, whenever your camera detects a shark in front of it (or an image of a shark :P)

How I built it

Libraries used

1. Google Cloud Vision API
2. Pillow
3. OpenCV
4. Numpy
5. Pandas

OpenCV opens up the camera then google cloud's vision API scans and senses objects present in the frame. These objects are stored in a Pandas data frame. If a shark is detected as an object in the frame, a fun fact about sharks will pop up where the shark has been detected in the frame.

Challenges I ran into

As this is my first Hackathon, I ran into a lot of challenges. A LOT
I had to learn how to use vision API as I don't have any experience with it.
The most difficult part was coming up with an idea for the project. I initially thought of making a classifier for shark species but I decided to do something new and made an AR project using vision API.

Accomplishments that I'm proud of

I'm proud of being a part of Major League Hacking. This being the first time I'm participating in a Hackathon (both offline or online). I got to learn vision API which is an extremely useful tool. I am also glad that I was able to complete this project during the hackathon. The community was really helpful and I also got a chance to interact with a lot of people.

What I learned

Vision API
Git and Github
Markdown
Project Management

What's next for Shark Bytes

This project currently uses stored images for facts which are limited. The number of fact images could be increased an stored in a database which can be integrated with python using sql. It also shows one image per run of the code, a key could be set to change the image anytime the user wants.