Skip to content

Latest commit

 

History

History
43 lines (33 loc) · 901 Bytes

README.md

File metadata and controls

43 lines (33 loc) · 901 Bytes

Visual-Perception

License: MIT

Visual Perception is a High Level Python Library to performs various task under visual recognition.

Currently, It Provides the Solution for Object Detection.

For Detection with pre-trained models it provides:

  • YOLOv4
  • TinyYOLOv4

Dependencies:

  • tensorflow >= 2.3.0
  • keras
  • opencv-python
  • numpy
  • pillow
  • matplotlib
  • pandas
  • scikit-learn
  • scikit-image
  • imgaug
  • labelme2coco
  • progressbar2
  • scipy
  • h5py
  • configobj

Image Object Detection Using YOLOv4

from visual_perception.Detection import Object_Detection

model = Object_Detection()
model.Det_YOLOv4()
model.Detect_From_Image(input_path='car.jpg',
                        output_path='./output_image.jpg')

from PIL import Image
Image.open('output_image.jpg')