High-Level Project Summary
Our project is to provide a new artistic way to explore the universe in the eyes of the previous or current artist. we developed a deep learning model to create an art of our world based on the user input from the NASA database and his favourite artist. it gives the people more passion for the space, especially since it's easier to watch, and more appealing to their taste of choice.
Link to Final Project
Link to Project "Demo"
Detailed Project Description
Our project mainly takes input from the user as words, consisting of which what he wants to see, in what artist they want to know about the universe, resulting in the art of this piece of information as if it were painted with his input artist.
- The project takes the input (text) and starts with Entity recognition, to extract the information about the artist, the planet or start he wants to see, and the year he wants the data in a range of it.
- Then we extract the information from both Nasa Database (using their API) to grab a picture of the requested info, and our Database to grab a picture of the style of the artist.
- Then we use a Deep learning model (Style Gans) to apply the style of the artist to the chosen content and feature it to use in an interactive website.
We hope to inspire more people to be more passionate about the space, and to make the space more accessible for everyone.
We used:
- Streamlit
- Tensorflow
- Spacy Model
- Deep learning models
- Python
- SQL
- API's
Space Agency Data
We used the data from Nasa Space images, this data was used to provide the user with what he wants to explore in our universe.
it inspired the project to add more reality to our artistic manner of showing the universe to the users.
Hackathon Journey
It was an extremely challenging and fun one, it was full of an adrenaline rush, going down, getting up and more of mixed feelings, but overall unique.
Time Management for most in the part, team work is always the key to success, learnt how to build very fast web applications too!
we all believed that art can change a lot of people's minds, always it can produce future enthusiastic scientists passionate about space, as technology evolves we thought of merging both to create a unique new thing.
developing the requirements into several small tasks, to make sure every feature is possible to be done, then thinking about how to merge some of them together, to make one better until all are merged together. (divide and conquer)
we mainly played a role char and changed everyone's places, so we can reach better decisions.
Our mentors, that always offered help, helped and encourages us to continue even though we almost gave up. thank you, beautiful people.
References
https://images.nasa.gov/
https://images.nasa.gov/docs/images.nasa.gov_api_docs.pdf
https://streamlit.io/
https://www.tensorflow.org/hub/tutorials/tf2_arbitrary_image_stylization
https://www.tensorflow.org/hub/tutorials/tf_hub_generative_image_module
Tags
#art #deep_learning #generation