Image Segmentation | DeepLabV3 ResNet50 | Train on Custom Dataset
🧱 Cracks Segmentation using DeepLabV3-ResNet50
📌 Overview
This video contains a complete pipeline for segmentation of cracks using DeepLabV3 with ResNet50 backbone, leveraging transfer learning with PyTorch.
The model is trained on a custom cracks dataset with binary labels (Background and Crack regions).
📋 What's Included?
✅ Data Preparation & Visualization
✅ DeepLabV3 Model Customization
✅ Crack Segmentation Dataset Loader
✅ Training & Validation with Loss Tracking
✅ Cosine Annealing LR Scheduler
✅ Inference & Prediction Visualization
🛠 Tech Stack
Python 🐍
PyTorch 🔥
Torchvision
Albumentations 🌈
Matplotlib 📊
OpenCV 📷
Google Colab ☁️
📥 Github Link : https://github.com/NitinCVOrbit/Crack...
#DeepLearning #ImageSegmentation #CrackDetection #PyTorch #DeepLabV3 #ResNet50 #ComputerVision #CustomDataset #SemanticSegmentation #ColabTutorial #MachineLearningProject