Creating a Sample App with Universe Projects

An example web application created with a Roboflow Universe dataset.

Written by Mohamed Traore

Last published at: June 1st, 2022

Roboflow Universe datasets can be used for project inspiration (use the search bar to look for any terms of interest!), gathering more data for a custom project, and even testing inference on trained Universe models.

Brad Dwyer, Roboflow CTO, created an example blackjack strategy web application that utilizes the Augmented Startups Playing Cards dataset and model from Roboflow Universe.

The app runs a playing cards object detection model in your web browser with roboflow.js (which is backed by tensorflow.js). Simply point your camera at a Blackjack hand, tell it what card the dealer has facing up, and it will tell you what your optimal next move is based on basic strategy. The code (and YouTube video, included below) are a great resource for creating your first computer vision apps.

🤳 Try it on your phone or in your web browser on GitHub Pages.


  • Try It - this project is deployed on GitHub Pages. You can try it on your phone or computer's webcam. Just point your camera at a Blackjack hand, select the dealer's face-up card, and it will tell you what to do.
  • Roboflow Universe - share computer vision datasets and pre-trained models,
  • Playing Cards Pre-Trained Model - shared on Roboflow Universe by Augmented Startups.
  • Roboflow YouTube - where you can follow along with the live-coding session of this app being built and find other computer vision content.
  • Roboflow - everything you need to create your own custom computer vision projects for free (if you're sharing them publicly). Upload, annotate, augment, train, and deploy in a single afternoon (then iterate and improve your model over time).

Next Steps

  1. Tell us about your project on the Show & Tell page on the Roboflow Community Forum!
  2. Contact us about getting featured on Roboflow Universe or the Roboflow blog and newsletter!
  3. Keep improving your dataset by utilizing Roboflow's pip package for Active Learning

This is a great place to receive project feedback and recognition from Roboflow users, and the Roboflow team. We like to select the best projects to highlight on Roboflow Universe's "Featured" section and in our weekly blog and newsletter releases.