We wanted to make a fun mystery game that you can race against with your friends!

What it does

Our game is story based. You first read about the basis of the game. The first level includes picking your character by dressing up or selecting the detective you most resemble. The second level includes finding objects at the crime scene and then correctly selecting the murder weapon. The third level includes hearing alibis for all the characters and questioning the characters with gaps or inconsistencies in their alibi. The last level is choosing who you think is the murderer.

How we built it

Google Cloud Technologies Used:

  • Firebase
  • Google Cloud Storage
  • Google Vision API
  • Google Assistant
  • DialogFlow
  • Google Cloud Functions
  • Google App Engine

Frontend: Our frontend was made with React and Ant Design for the design library.

Image Upload API: The image upload api was made using Express and Multer using Node.js along with Google Cloud Storage to temporarily store images in Firebase.

Vision API: The image investigation API was made using Google Vision API to search for labels and objects unique to each detective. The image comparison API was made using Deep AI to find the most likely detective match. These APIs were made using Flask in Python.

Google Assistant: The smart agent was created using DialogFlow and Google Cloud Function written in Node.js.

DevOps: The Vision API was deployed on Google App Engine and the frontend was deployed using Firebase.

Challenges we ran into

One of the challenges we ran into was finding an API that would most effectively match users to detectives. After a lot of research, we found that the best way to do this would be to use three different computer vision endpoints and combine them into one. Another challenge we came across was how to handle image uploads since none of us have ever worked with it before. After reading about Cloud Storage and watching a couple tutorials, we were able to create an API that handles image upload and processing in a clean and fast method.

Accomplishments that we're proud of

We are proud of creating a finished product within 48 hours complete with a frontend, two backend APIs and a smart home assistant.

What we learned

We learnt how to use Google Cloud Storage and Google Vision API since this was our first time using them. We also learnt how to effectively make a single page but multi state React web application.

What's next for Murder Mystery

We would definitely add more levels to give more clues to the user and make the game longer. We would also automatically generate different culprits each time so the game is replayable.

Built With

Share this project: