andfanilo/streamlit-drawable-canvas

Do you like Quick, Draw? Well what if you could train/predict doodles drawn inside Streamlit? Also draws lines, circles and boxes over background images for annotation.

What it solves

It provides a way to integrate an interactive sketching canvas into Streamlit applications, allowing users to draw, annotate, and manipulate shapes directly within a web app without needing to build a custom frontend component from scratch.

How it works

Built as a Streamlit component using Fabric.js, it renders a canvas in the browser that supports various drawing modes (freedraw, lines, rectangles, circles, polygons, and text). It can be initialized with a background image or color and sends the resulting image data (as a numpy array) or object properties (as JSON) back to the Streamlit backend for further processing.

Who it’s for

Developers building Streamlit apps who need user-input for visual data, such as image annotation tools, sketching interfaces, or simple graphic editors.

Highlights

  • Multiple Drawing Modes: Supports freehand drawing, points, lines, rectangles, circles, polygons, and text.
  • Object Manipulation: Allows users to rotate, skew, scale, and move objects already placed on the canvas.
  • Data Integration: Returns both the raw image data and detailed JSON properties of every drawn object to the Python backend.
  • Annotation Support: Includes a specific labeled_rect mode for creating labeled bounding boxes.
  • Customizable Backgrounds: Supports background images with options to fit via stretching or containing.

Related

  • Project
    t8/hypura

    A storage-tier-aware LLM inference scheduler for Apple Silicon that enables running models larger than physical memory by streaming tensors from NVMe storage.

  • Project
    nest/nest-simulator

    A scalable simulator for spiking neural network models used to study the dynamics, size, and structure of neural systems.

  • Project
    fatiando/verde

    A Python library for processing and gridding spatial data using a machine-learning-inspired interface similar to scikit-learn.

  • Project
    lablup/backend.ai

    A container-based computing cluster platform that orchestrates ML frameworks and provides multi-tenant access to heterogeneous hardware accelerators like GPUs and NPUs.