Pytorch implementation of the U-Net for image semantic segmentation, with dense CRF post-processing
-
Updated
Feb 7, 2019 - Python
Pytorch implementation of the U-Net for image semantic segmentation, with dense CRF post-processing
Convolutional Neural Networks: (1) based on UNet; (2) FCN8 for Image Segmentation of Pascal VOC 2012 dataset written as part of my MSc in Artificial Intelligence degree. Written in Tensorflow 2.0 with Keras Functional API.
UNet binary segmentation - inference software (on normal images)
In this repository you can find the jupyter notebooks used to take part at the competitions created for the Artifical Neural Networks and Deep Learning exam at Politecnico di Milano.
This repository accompanies the publication "U-Net based particle localization in granular experiments: Accuracy limits and optimization," published in Granular Matter (2026). It contains the U-Net implementation, pretrained weights, labeled datasets, Jupyter notebooks, and instructions for reproducing the results.
this repo's goal is an improvement in overall development capability about image processing
Segmentation of aerial images with Deep Learning.
Anatomy-constrained chest X-ray pipeline: lung segmentation + two-stage heatmap regression for hemidiaphragm landmark localisation.
Computer vision portfolio: CIFAR-10 classification with ResNet50, steel defect segmentation with U-Net, and interactive Streamlit dashboards
YOLOv8 object detection + U-Net semantic segmentation for waste classification. Features custom and transfer learning implementations with 73% performance improvement on TACO dataset.
This project implements a GAN model for converting grayscale images to color.
Blood Vessel Segmentation was done on Messidor Dataset. Using the weights of a model which was trained on Drive2004 Dataset and ChaseDB
AI-Driven car announcement helper
A decentralized, diffusion-based U-Net framework for privacy-preserving brain tumor segmentation from MRI images.
UNet that cross link features.
This is a Deep Learning project that aims to classify brain MRI's as either containing a tumor or not.
Multi-class 3D protein complex segmentation in cryo-electron tomography using MONAI, FlexibleUNet, and W&B.
Semantic segmentation with a compact U-Net — per-pixel defect/lesion masks with Dice/IoU, beating an intensity-threshold baseline (PyTorch).
Brain Tumor MRI Segmentation using U-Net with EfficientNetB3 backbone | Transfer Learning | Streamlit Web App | TensorFlow/Keras
To associate your repository with the unet topic, visit your repo's landing page and select "manage topics."