devdatta95 / object_detection_using_yolo_and_ssd

Created a Flask application that can be used to test and deploy Object Detection Model on Images and Videos

Realtime Obejct Recognition Using YOLO and SSD

Realtime object recognition using the OpenCV 3.3 dnn module + pretrained MobileNetSSD caffemodel.

Installation

All the dependencies can be installed using pip. Just use the following command from the root directory of the project.

pip3 install -r requirements.txt

How to run this script?

There are two options for video source:

• Webcam
• Android device running IP Camera (https://play.google.com/store/apps/details?id=com.pas.webcam&hl=en)

To run the script using webcam as source :

python3 real_time_object_detection.py  --source webcam

To run the script using IP Webcam as source, open the real_time_object_detection.py and edit the following line to match your host :

host = 'http://192.168.0.101:8080/'

Then to run the script using IP as source :

python3 real_time_object_detection.py  --source web

For any questions, create an issue in this repository.