Object detection web application developed using Detectron2 pre-trained model
-
Create a new conda or virtual environment to run this project (recommended if you don't want to mess up with the libraries in existing environment)
-
Activate or switch to the environment and clone the repository
git clone https://github.com/veeruamma/Detectron2-Object-Detection.git
- Install the required libraries using the following command
pip install -r requirements.txt
- Install the detectron2 manually by referring to following command (it's commented in requirements.txt)
pip install 'git+https://github.com/facebookresearch/detectron2.git'
- After the successfull installation, run the file clientApp.py
python clientApp.py
- Application starts running on localhost at port 5000 as shown below
- The landing page can be viewed as
- Obejct Detection Output