roboflow / streamlit-web-app

A web-based application for testing models trained with Roboflow. Powered by Streamlit.

Roboflow: Streamlit Computer Vision App for Web Browser Deployment

A web-based application for testing models trained with Roboflow. Powered by Streamlit.

🎬 videos

Every week we create tutorials showing you the best new updates in Computer Vision. 🔥 Subscribe, and stay up to date with our latest YouTube videos!

How it Works:

💻 run locally

Complete these steps with your console/terminal

# clone the repository
git clone https://github.com/roboflow/streamlit-web-app.git

# navigate to the root directory
cd streamlit-web-app
cd streamlit

# set up your python environment and activate it
python3 -m venv venv
source venv/bin/activate

# install the requirements
pip install -r requirements.txt

# run the app
streamlit run Home.py

Troubleshooting:

• For Mac users: be sure that you have Homebrew installed.
• Unable to resolve wheel for av or aiortc packages: Install pkg-config by executing brew install pkg-config in your Terminal
• If you wish to process video streams with Streamlit apps, be sure to also have ffmpeg installed: after installing Homebrew, execute brew install ffmpeg in your Terminal

-- Check here for more on ffmpeg installation: https://github.com/roboflow/video-inference#requirements

• Ensure that you have opencv-python-headless installed in your environment, instead of opencv-python

• Unable to install the av package with pip? Try executing conda install av -c conda-forge in your Terminal

-- Note: Anaconda or Miniconda is required for this method. Be sure that conda-forge has been made available in your channels. https://conda-forge.org/docs/user/introduction.html#how-can-i-install-packages-from-conda-forge

Next Steps:

1. Customize the app further for your use case: https://docs.streamlit.io
2. Create another custom app or solution with your model
3. Create new models. Need inspiration? Try searching Roboflow Universe
4. Getting false detections? Model not performing to your expectations? Try Active Learning (Smart Sampling) and improving your Dataset Health:

• Rapid Testing and Deployment Iteration with Roboflow Train and the Deploy Tab
• How to Use Dataset Health Check to Improve Model Quality
• What is Active Learning? | Why You Should Implement Active Learning
  • Active Learning with Roboflow's Python Package: https://docs.roboflow.com/python/active-learning

Increase Your Computer Vision Knowledge:

• https://roboflow.com/learn

Object Detection

Image Classification

[Instance] Segmentation

Object Detection vs. Image Classification vs. Keypoint Detection

Semantic Segmentation vs. Instance Segmentation: Explained

🐞 bugs & 🦸 contribution

If you notice that any of the code is not working properly, create a bug report and let us know.

If you have an idea for new functionality we should do, create a feature request. If you feel up to the task and want to create a Pull Request yourself, please feel free to do so.

We are here for you, so don't hesitate to reach out here, or on our Community Forum.