-
Captured a butterfly!
-
back end, untested photos.
-
Uploading an image on a OnePlus android phone
Inspiration
After being stuck indoors, we wanted a non-social way to stay in contact with nature and notice the little details around us. We wanted to bring awareness and appreciation to the wildlife in our community.
What it does
Using the latest visual recognition techniques, users can upload pictures of animals they take in the real world. The pocket index, or Pokedex, will store the image, label the animal's species, award points, and gives a tad bit of info about the animal you discovered.
How we built it
A website that is designed to run on mobile will allow users to upload an image. The image will be sent to the IBM Watson API which will returns what Watson thinks the image is with values of certainty. This data is then processed and the results are displayed on the website.
Challenges we ran into
We found it particularly difficult to integrate the API to the website created.
Accomplishments that we are proud of
Using machine learning to train a model to recognize plastic on hand insects.
What we learned
The applications of visual recognition and how far its been developed. Learning how to use a serverless server was also new and interesting.
What's next for Real Life Pokedex using Visual Recognition
With more time, we want to add a location component where users have the option to share the location of their animal to a collective google map.
Looking to be considered for Random hack & Sustainability hack
Log in or sign up for Devpost to join the conversation.