Falldown detection

Using the YOLOv8n model to detect changes in the length-to-width ratio of a person’s bounding box (bbox) to determine whether a fall has occurred can assist hospitals or nursing homes in monitoring elderly individuals for falls, thereby reducing the workload of caregivers.

Hardware prepare

reComputer AI Box	reComputer Industry AI Box
Purchase Now	Purchase Now

Software prepare

Install the runtime environment

sudo apt update && sudo apt full-upgrade -y && sudo apt install hailo-all

Install the project

git clone https://github.com/Seeed-Projects/fall_detection_with_AIBox.git
cd fall_detection_with_AIBox

Prepare the python environment

python -m venv .env --system-site-packages  && source .env/bin/activate
pip install -r requirements.txt

Run the project

python app.py -i ./falldown_test.mp4 -n ./yolov8n.hef --show-fps -l ./common/coco.txt

Result

Reference

This project references Hailo-Application-Code-Examples.