Real-Time-Object-Detection-Recognition
Code for realtime object detection and recognition . The model is trained on caffe and uses OpenCV. The code and it's testing was done on a CPU .
Getting Started
- Download MobileNetSSD_deploy.caffemodel and MobileNetSSD_deploy.prototxt.txt for real time object detection.
- Host all the files on the same directory
- It is always a good practice to create a virtual environment for every project , not compuslory but ensures that dependencies are not hampered elsewhere.
- It contains more than around 40+ objects trained model.
Installing
pip install -r requirements.txt
Deployment
- Tested on
- Intel i3 6006U CPU
- OS : Linux Fedora 28
- 4GB RAM
Best Results based on number of objects detected
To Run : python objdet.py -p [PATH TO PROTOTXT FILE] -m [PATH TO CAFFEMODEL DEPLOY] -c [CONFIDENCE]
Contributing
If you have any :
- Pull Requests : make one, work is always appreciated.
- Issues : they are what help us improve, do create if any.
- Idea : feature requests etc.
- Suggestions : The world of Open Source
Authors
- Ashwin Phadke
License
This project is licensed under the MIT License - see the LICENSE.md file for details
Acknowledgments
- Amazing Ad.
- Open Source Community.
- Last mile contributors.