Building interactive applications for image data

on YouTube ↗29m

" summary

Live demos show Dash apps with drawing tools that emit JSON geometry, annotations saved to a store and wired to callbacks; the UI integrates with plotly graphs and a soon-to-ship drawing mode. Image processing uses scikit-image for preprocessing, line profiles, superpixel segmentation and feature extraction, then a scikit-learn random forest classifies foreground/background to remove backgrounds. Code is reusable, Python-only, and the demos consistently produce usable segmentations.

AI-written summary · about

" chapters (10)

1. 0:00Start
2. 1:04Image processing for science and business
3. 3:02Interaction with images annotations
4. 8:08Available components
5. 16:48The Scientific Python ecosystem
6. 20:58A toolbox for manipulating images
7. 23:17Advanced algorithms
8. 24:27Extracting features
9. 25:05Remove image background
10. 27:49Getting started: finding documentation

" related talks

• 41mDash: data exploration web apps in pure PythonChelsea Douglas · PyData DC 2018
• 26mDeveloping Dashboard Applications using BokehLuke Canavan · PyData DC 2018
• 32mDash: Interactive Data Visualization Web Apps with no JavascriptDom Weldon · EuroPython 2019
• 27mUsing Dash to Create Interactive Web Apps for Non-Technical AudiencesJoseph Willi · PyOhio 2019
• 02h 14mData Applications with BokehBryan Van de Ven · PyData San Francisco 2016
• 29mUsing Dash by Plotly for Interactive Visualisation of Crime DataLeo Broska · PyCon AU 2019
• 30mMeta-Dashboards: Accelerating Geospatial Web Apps Creation with VoilàDavide De Marchi · PyData Paris 2025
• 45mDeveloping Dashboard Applications Using BokehBryan Van de Ven · PyData Los Angeles 2018