Bird Vision

Quickly scan a bird or upload a photo, and see what species it is! You can see information about the bird, as well as if it's endangered and information about climate change, using an AI chat-bot.

Comment

Inspiration

Me, Daniel, and Alvish love the outdoors and learning more about nature. When we saw the prompt about inspiring wildlife conservation efforts, we were reminded of how much we love learning more about the different plants and animals in the world. In particular, I had recently come across an app which lets you scan photos of plants and see what species they are. We believe that in learning more about the environment around you, you feel more connected to it and become more likely to make efforts to protect it.

What it does

Bird Vision is a web app that allows you to take a picture of a bird, or upload a photo. Our home-trained AI/ML model analyses the photo and predicts which type of bird it is. From there, you can learn more about whether or not the species is endangered, how species endangerment is connected to climate change, and what you can do to conserve wildlife.

How we built it

We built Bird Vision by applying two different AI models. First, we made a completely custom AI model using Teachable Machine, which can take in an image of a bird, and identify it using our extensive dataset of birds. Secondly, we incorporated Gemini as a chat-bot which can interact with the user and open up a discussion. These technologies are hosted on a website running on Django.

Challenges we ran into

We found it quite challenging to navigate AI technologies that we were not previously familiar with, and dealing with the specifics of fine-tuning our model to work just right.

Accomplishments that we're proud of

We're really proud of the ease-of-use of our website. I believe in accessibility, and I think that the friendliness of our website could inspire people to go out in nature--- especially children---and learn more about our world.

What we learned

We all learned a lot about AI/ML. For myself, at least, this was my first big glimpse into working with AI in a larger project. But what I will say for everyone is that we learned a lot about the importance of teamwork and resilience. We encountered disagreements in terms of design decisions and aesthetics, and many challenges with things not working and stress of time pressure. But I'm proud of the way we pushed through them as a team and managed to create something great.

What's next for Bird Vision

Bird Vision still has much room to grow, and we're happy to expand it. For one, the AI/ML model is only trained on the smaller version of our dataset. Given more time and processing power, we could create a model that is not only more accurate, but more expansive -- covering not just birds, but all animals and plants. Something that would really improve the look and feel of Bird Vision would be to port it to a native mobile app. I think that Bird Vision is something that is something that I'd love to share with more people on the iOS and Android stores.

Built With

Share this project:

Updates