Skip to content

This repo uses pre-trained SSD MobileNet V3 model to detect objects belonging to 80 different classes in images and videos

Notifications You must be signed in to change notification settings

zafarRehan/object_detection_COCO

Repository files navigation

Object Detection

This repository uses pre-trained MobileNet-SSD v3 model for Object Detection https://github.com/opencv/opencv/wiki/TensorFlow-Object-Detection-API

The model uses COCO dataset for training https://cocodataset.org/#overview

The COCO dataset consists of 80 classes of images (View all) so the objects detected would belong from one of these classes only.

You can download the test_video from:
https://drive.google.com/file/d/1EPngtG-X4NCOoadeo1Fk556Yhejuvp5C/view?usp=sharing
or you can use any video of your choice

Required Libraries

pip install opencv-python

Usage

This repository has 2 files:
1. detection_code_image.py
2. detection_code_video.py

Image

Change the 'your image path' in detection_code_image.py at line:23 to the path of your image to detect objects in your Image

Example Input Image:

Example Output Image:

Video

Change the 'your image path' in detection_code_video.py at line: 21 to the path of your video to detect objects in your Video

Example Input Video:

Example Output Video:

About

This repo uses pre-trained SSD MobileNet V3 model to detect objects belonging to 80 different classes in images and videos

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages