Overview
Learn to implement semantic image segmentation using U-NET architecture from scratch in this comprehensive 52-minute PyTorch tutorial. Dive deep into the U-Net implementation, dataset preparation, training process, and utility functions. Follow along as the instructor builds a complete image segmentation pipeline using the Carvana Image Masking Challenge dataset. Gain hands-on experience in creating custom datasets, designing the U-Net architecture, setting up training loops, and evaluating model performance. Perfect for deep learning enthusiasts looking to master advanced computer vision techniques and understand the intricacies of image segmentation tasks.
Syllabus
- Introduction
- Model from scratch
- Dataset from scratch
- Training from scratch
- Utils almost from scratch
- Evaluation and Ending
Taught by
Aladdin Persson