https://github.com/abhik-chakraborty/face-detection

https://github.com/abhik-chakraborty/face-detection

Last synced: 12 days ago
JSON representation

• Host: GitHub
• URL: https://github.com/abhik-chakraborty/face-detection
• Owner: Abhik-Chakraborty
• Created: 2023-03-19T20:53:30.000Z (almost 2 years ago)
• Default Branch: main
• Last Pushed: 2024-01-30T06:10:30.000Z (11 months ago)
• Last Synced: 2024-01-30T07:27:01.761Z (11 months ago)
• Language: Python
• Size: 28.5 MB
• Stars: 0
• Watchers: 1
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

        
# CV-object-detection

This repository includes the following scripts:

* license_plate_detection.py - Detect and read license plates in a video file

* object_detection.py - Detect objects in a video file

* face_detection.py - Recognize faces in a video file

## Read License Plates

This script reads the video file, looks for license plates and if found, write the image of the frame 

with bounding boxes and also a csv with list of plates detected and their corresponding images

cmd to run

```console 

python license_plate_detection.py --input-file-path  --output-dir 

```

Example output

![Example Output](example_outputs/license-plate.png)

## Detect Objects

This script reads the video file, looks for objects and if found, write the image of the frame 

with bounding boxes

The script uses pretrained mobilenet model. The model weights and graph are in mobilenet_model folder

cmd to run

```console 

python object_detection.py --input-file-path  --output-dir  --model-weights-path 'mobilenet_model/frozen_inference_graph.pb' --model-graph-path 'mobilenet_model/graph.pbtxt'

```

Example output

![Example Output](example_outputs/object-detection.png)

## Recognize faces

This script reads the video file, looks for faces, compares the faces with known faces and if there is a match then 

writes an image with bounding boxes and the filename of the known face image

The script uses [face-recognition python library](https://pypi.org/project/face-recognition/) 

Known persons directory should have images with filname in the format {person_name}.{format}

cmd to run

```console 

python face_detection.py --input-file-path  --known-persons-dir ' --output-dir 

```

Example output

![Example Output](example_outputs/face-recognition.png)