Beun Team Twente 3.1-beta-rc2.1| Putting the 'Art' in Artemis
Awards & Nominations
Beun Team Twente 3.1-beta-rc2.1 has received the following awards and nominations. Way to go!
Putting the 'Art' in Artemis
Spaceship Builder
Summary
Using computer vision algorithms, children’s colored spaceships are imported in an animation that shows their starship on a voyage from the Earth to Mars on their own smart-phone or computer! The drawings can be scanned by the parents using their smartphone camera. Our solution to the challenge inspires others about humanity’s road to Mars in a creative and accessible way, by allowing everyone to put the ‘Art’ in Artemis!
How We Addressed This Challenge
The challenge was about creating an artistic work to inspire others about humanity's road to mars. With our take on this challenge we would like to inspire children about space exploration and going towards mars. In our opinion there is no better way to inspire children than them seeing their own creations go to space. With our project they can do just that.
Just print the spaceship and let them colour it in. Of course our program also supports digitally colored images to save on paper. After the spaceship is colored in, the images can be uploaded and in just a few seconds the spaceship will start the journey towards mars. The back-end of the program also supports the ability to show other recently uploaded work in the background. This means that it is also possible to show friends and family to see each other's rockets when launching, also inspiring them. The program can also be used in elementary schools and aid with lessons about space. All in all, our solution to the challenge inspires others about humanity’s road to Mars in a creative and accessible way, by allowing everyone to put the ‘Art’ in Artemis!
How We Developed This Project
We divided the project into several smaller projects. Some of our group members were skilled in using computer vision algorithms and some in server development. The other part of the group focused on getting a website and being able to show the animations on the screen.
We used OpenCV libraries in python to automatically obtain the images of the space rockets and applied projective transformation such that images of the rocket could be taken under an angle and still would result in a nice image to show. We achieved this by matching the colored rocket with a reference image using the SIFT algorithm (included in OpenCV libraries). A visualization of the matching is shown in the image below. We then removed the background such that only the colored rocket will be seen on the screen. Further processing also included obtaining the face of the astronaut from the window of the space rocket.
These algorithms were called from the Flask server upon a new photo upload. All the image processing was performed in-memory so that the app is quick and scalable, because no Disk I/O was needed. After the images were processed, they are stored in a database (so that we can have multiple people land on Mars - which was not implemented in the front-end due to the time constraints of the Hackathon). The server upload endpoint, returns the processed photo of the rocket and the astronaut (in base64 format), alongside the drawings and locations of other rockets which have already landed on Mars. The base64 assets can then be used by the JS front-end which runs on the user’s browser.
Using HTML5’s canvas element in combination with the PaperJS (paperjs.org) library, the spaceship is animated in its voyage to mars on the webpage. Keyframes are used to time the animation. There is no need to install any software as the application is completely accessible from a browser. The vector image artwork of the Earth and Mars, as well as the rocket and astronaut drawing, has been created by our group. The audio added to the animation is from NASA archives.
And of course, we had fun designing and colouring some of the rockets ourselves. You know... For testing purposes...
How We Used Space Agency Data in This Project
The platform is meant for creative and artistic purposes that can inspire and even educate kids, therefore using realistic NASA data significantly contributes to the experience. For the demo project, we used audio files from the official collection of NASA, but the web-app can be extended with more audio-visual media from other space agencies as well
Project Demo
The following diagram shows how the process of how our project is used:
You can download the rocket template from the website. Once you print and colour it, you can use the website with your phone to upload a photo. Once the processing is complete, your rocket will be ready on the launchpad and the countdown begins.
You’ll see your rocket travelling through space and slowing down before landing on Mars. Once you land there, you can enjoy the scenery of the red planet.
A small demonstration of the website in action can be seen in this video: https://youtu.be/qBRNghbNuaM (this is not the final video).
Data & Resources
The audio files from NASA were used throughout the platform. Below are the links where they are found:
- Countdown, Liftoff and Landing sound: https://www.nasa.gov/mp3/640392main_STS-26_Liftoff.mp3
- Sputnik Beeps used during travel time: https://www.nasa.gov/mp3/578626main_sputnik-beep.mp3
The other assets, such as images and the drawing templates, were designed by our team. The drawing template can be found here:
https://github.com/Beun-Team-Twente/spaceapps2020/blob/master/static/download/Rocket.pdf