Instance Segmentation with Detectron2 - Custom Dataset Training and Testing

Learn how to build a custom instance segmentation model using Detectron2 for photo segmentation. Master the process of annotating images, training the model on both Windows and Linux environments, and implementing inference code to test the model on new images. Gain hands-on experience with deep learning techniques for achieving high-performance panoptic segmentation results on your own dataset. Follow along as the tutorial covers image annotation, model training on CPU and GPU, and practical testing methods to apply your custom-trained model to fresh images.

Introduction
Annotate the images & image discovery
Train the model on Windows CPU
Train the model on Ubuntu GPU
Test the model