Image Segmentation | FCN ResNet50 | Train on Custom Dataset
🚗 Car Image Segmentation using FCN-ResNet50 | Segmentation with PyTorch
📌 About the Project
In this video, I walk you through a complete pipeline for segmentation of cars using FCN (Fully Convolutional Network) with a ResNet50 backbone, powered by transfer learning in PyTorch.
This project uses a custom car segmentation dataset with classes ("Car", "Wheel", "Lights", "Window"), and demonstrates data loading, model training, and prediction visualization.
📋 Key Features
✅ Data Preparation & Augmentation
✅ FCN-ResNet50 Model Setup with Transfer Learning
✅ Custom Dataset Class for Car Masks
✅ Inference with Visualization of Segmentation Masks
🛠 Tech Stack
Python 🐍
PyTorch 🔥
Torchvision
Albumentations 🌈
OpenCV 📷
Matplotlib 📊
Google Colab ☁️
📁 Download Dataset & Code
Get the complete code and dataset from my GitHub repository:
🔗 https://github.com/NitinCVOrbit/Car_I...
👍 Like, Share & Subscribe if you found this helpful!
💬 Drop your questions in the comments – I’d love to help!
#CarSegmentation #SemanticSegmentation #PyTorch #DeepLearning #ComputerVision #FCN #ResNet50 #ImageSegmentation