Ultimate detection and recognition of predefined objects in real time.
An object is identified and recognized whenever it comes to the frame. The detected object is marked within a rectangle with label specifying what object it is.
Objects recognized:
- Face
- Eye
- Smile
- Numberplate
- Wallclock
Follow the steps below:
- Clone or download video.py, common.py and ObjectRecognition.py.
- Keep video.py and common.py files in the same folder as ObjectRecognition.py.
- Make sure to download haar folder which contains all the haar-cascade files required for object recognition.
- Also make sure that you give the path correctly for the cascade files in args.get function within ObjectRecognition.py - Change the path (provided) and make it correspond to the path for the haar-cascade files on your computer.
- Run ObjectRecognition.py - This opens your webcam and the objects (specified above) are recognized.
Feel free to make modifications.