Deep Learning

Deep learning study notes

Image Segmentation with Mask R-CNN

Mask R-CNN with OpenCV

You can automatically segment and construct pixel-wise masks for every object in an image, thereby to segment the foreground object from the background. We just need perform instance segmentation using the Mask-RCNN architecture.

Difference between semantic segmentation and instance segmentation

Semantic Segmentation: require us to associate every pixel in an input image with a class label (including background). Instance Segmentation: compute a pixel-wise mask for every object in the image, even if the objects in the same class.

The Mask R-CNN we are using here were trained on the COCO dataset, which has L=90, thus the resulting volumn size from the mask module of the Mask R-CNN is 1009015*15

Configuration

• mask-rcnn-coco/ : The Mask R-CNN model files
  • frozen_inference_graph.pb: The Mask R-CNN model weights. The weights are pre-trained on the COCO dataset.
  • mask_rcnn_inception_v2_coco_2018_01_28.pbtxt: The Mask R-CNN model configuration.
  • object_detection_classes_coco.txt : All 90 classes are listed in this text file, one per line. Open it in a text editor to see what objects our model can recognize.
  • colors.txt : This text file contains colors to randomly assign to objects found in the image.

RNN

1. Udemy: Deep Learning A-Z™: Hands-On Artificial Neural Networks

Dataset

SuperDataScience Deep Learning A-Z™: Download Practice Datasets(https://www.superdatascience.com/pages/deep-learning)