Deep learning libs to process images and videos and execute tasks like segmentation, classification and etc… Pixelib now is using tensorflow and pytorch as a core so let’s do some tests using the pytorch version.
1 – Installing the libs
pip3 install torch==1.8.2+cu102 torchvision==0.9.2+cu102 torchaudio==0.8.2 -f https://download.pytorch.org/whl/lts/1.8/torch_lts.html
pip3 install pycocotools
pip3 install pixellib
2 – Importing the libs and downloading an image as example:
import numpy as np
import cv2
from google.colab.patches import cv2_imshow
!wget https://github.com/ayoolaolafenwa/PixelLib/blob/master/Images/github.jpg?raw=true -q -O input.jpg
!wget https://github.com/ayoolaolafenwa/PixelLib/releases/download/0.2.0/pointrend_resnet50.pkl -q -O pointrend_resnet50.pkl
im = cv2.imread("./input.jpg")
cv2_imshow(im)
3 – Creating a segmentation using the pre-trained model pointrend_resnet50.pkl:
import pixellib
from pixellib.torchbackend.instance import instanceSegmentation
ins = instanceSegmentation()
ins.load_model("pointrend_resnet50.pkl")
ins.segmentImage("./input.jpg", show_bboxes=True, output_image_name="./output_image.jpg")
im = cv2.imread("./output_image.jpg")
cv2_imshow(im)
4 – Let’s check the results:
5 – Conclusion
Pixelib showed great results make segmentations! Let’s try to do some custom models next time!
Reference: https://github.com/ayoolaolafenwa/PixelLib
English