- Footfall analysis in real-time (supporting multiple video streams e.g., IP cameras, webcam) with a desktop app.
- Calculate demographics such as dwell times/heat maps at a particular location e.g., near shelfs.
- Customize the layout e.g., shelf control.
- People detection (using YOLO) along with age/gender detection if applicable.
- Track the path of customers in the stores.
- Analysis data is also stored in the logs.
Example test run showcasing multiple video streams with customer detection/tracking, dwell times/heat maps:
Example test run showcasing shelf control:
NOTE: This project is experimental and not maintained actively. If any bugs/problems are encountered, please open an issue.
TBD!
First up, install all the required Python dependencies by running: pip install -r requirements.txt
NOTE: Development environment is Windows/Python version 3.7.4 (there can always be version conflicts between the dependencies, OS, hardware etc.).
The desktop application is powered by PyQt, run it with: python footfall.py
.
saimj7/ 14-06-2020 - © Sai_Mj.