Skip to content

Latest commit

 

History

History
118 lines (66 loc) · 2.9 KB

README.md

File metadata and controls

118 lines (66 loc) · 2.9 KB
Raw

Pete-tionary

Play at https://pictionary.pbatch.net/.

Welcome to Pete-tionary, a combination of image synthesis and telephone pictionary.

robot

This is a fun game played in groups which proceeds as follows:

Each player writes down a prompt.

dog

Each player selects the image that best matches their prompt.

dog_single

Each player passes their image on and receives a new image from another player.

death_single

Each player must then try to write a prompt that generates a similar image. The game continues in this fashion until each player has seen each initial image once.

supper

Once the game ends, you can scroll through the stories you have made.

dog_final

death_final

IMPORTANT! If the website breaks while you are playing, perform a hard reset by deleting the local storage.

Example Game

banana_penis

midnight_blood_ritual_cult

popworld

spiralling_out_of_control

the_greem_army_from_toystory

Credits

Thanks to Jens Goldberg for creating the Direct Visions notebook.

Special mention to Phil Wang for creating the Deep Daze repository, and to Ryan Murdock for the discovery of the technique itself.

A huge of amount of gratitude to those who have given their time to test the game at various points.

TODOS

  • Connect Redux state with Amplify
  • Make the S3 URL checks a graphQL subscription (instead of pinging from Lambda)
  • Add the current queue size to the waiting screen
  • Run the server from inside Google Colab to save money

Deploying your own version of the website

  1. Install AWS CDK and the AWS CLI

  2. Make a private ECR repository

  3. Follow the commands in the ECR repository to upload the Docker image

  4. Set the global variables in backend/pictionary/pictionary_stack.py

  5. Deploy the CloudFormation stack (run this command in backend)

    cdk deploy --outputs-file "../frontend/src/constants/cdk.json"

  6. Build the website (run this command in frontend)

    npm run build

  7. Deploy the CloudFormation stack again (See step 4)

  8. Start the server

    python server.py

Contact

If you have any questions, queries or concerns I would love to hear from you.

Please email me at peter_batchelor1@hotmail.com.