AniDex

Inspiration:

The inspiration behind our project, Anidex, stemmed from the desire to merge technology with wildlife exploration and education. We recognized a gap in the market for a gamified platform that not only entertains users but also fosters a deeper understanding and appreciation for wildlife. Our team shares a passion for nature and conservation, and we believe that by gamifying the experience of uncovering wildlife, we can inspire more people to engage with and care about the natural world.

In today's fast-paced world, people are becoming increasingly disconnected from nature. Through Anidex, we aim to bridge this gap by providing a fun and educational platform that encourages users to explore their surroundings and discover the diverse range of animal species that inhabit our planet. We believe that by leveraging generative AI, we can make the process of identifying and learning about wildlife more accessible and engaging for users of all ages.

In the context of generative AI and sustainable development goals, Anidex aligns with the goal of promoting environmental awareness and conservation efforts. By fostering a sense of curiosity and connection to the natural world, we hope to inspire positive action towards protecting and preserving our planet's biodiversity.

What it does

It allows users to capture images of animals nearby then using Google Gemini's Vector AI image-to-text processor converts the image into text format then back to image format through base64 in Pixelated format. The animals in our collection is stored in our firestore database where we're able to fetch and a ui of each animal is includes a chatbot where users can interact and learn more of said animal.

How we built it

We built using React.js and TailwindCSS on the front-end as for a more complicated back-end it uses Node.js and Express.js to call Vector AI API to utilize Gemini 1.0 for our image processing. The database used is firebase / firestore. We deployed it using Google Cloud.

Challenges we ran into

The toughest challenge for us is to convert images into base64 format within a file size limit to parse into our Gemini API to produce a text back then converting the text back into image. This whole process was the most difficult. Additionally, we ran into some challenges with prompting our model to give tailored service to each animal.

Accomplishments that we're proud of

We are proud to have a complete product that is usable and has many use cases. We are proud of our team's combined effort through sleepless nights within the last two days to accomplish this task. It is also most of us first-time using AI which we are incredibly proud of what we managed to pull off.