Thomas Brox, Philipp Fischer, Olaf Ronneberger - 2015
Publications: arXiv Add/Edit
There is large consent that successful training of deep networks requires many thousand annotated training samples. In this paper, we present a network and training strategy that relies on the strong use of data augmentation to use the available annotated samples more efficiently. The architecture consists of a contracting path to capture context and a symmetric expanding path that enables precise localization. We show that such a network can be trained end-to-end from very few images and outperforms the prior best method (a sliding-window convolutional network) on the ISBI challenge for segmentation of neuronal structures in electron microscopic stacks. Using the same network trained on transmitted light microscopy images (phase contrast and DIC) we won the ISBI cell tracking challenge 2015 in these categories by a large margin. Moreover, the network is fast. Segmentation of a 512x512 image takes less than a second on a recent GPU. The full implementation (based on Caffe) and the trained networks are available at http://lmb.informatik.uni-freiburg.de/people/ronneber/u-net .
Implementation of https://arxiv.org/pdf/1505.04597.pdf
Segmentation of satellite images to map individual water bodies or forest areas for enhanced remote monitoring.
Early Barette Cancer detection is a project for the practical course machine learning in medical imaging uses U-Nets to segment the images and find out the traces of cancerous region
Satellite image segmentation for the Airbus Ship Detection Challenge organized by Kaggle (https://www.kaggle.com/c/airbus-ship-detection). This project was created as the project assignment for the course Deep Learning in Practice with Python and LUA (VITMAV45).
Ho Chi Minh is designed to extract textual information from tables presented in PDF, pictures or other format. Хошимин предназначен для извлечения текстовой информации из таблиц, представленных в PDF, картинках или ином формате.
Neural network comparison for Kaggle Ultrasound Segmentation competition
This project was developed for identifying vehicles in a video stream. The project is a corner stone for a real time vehicle tracking algorithm that employ semantic pixel-wise methods. This project solves the tracking problem for the Udacity final project in a different way that the general approach presented in the course. Instead of using the HOG features and other features extracted from the color space of the images, we used the U-Net which is a convolutional network for biomedical image segmentation.
The following is a new architecture for robust segmentation. It may perform better than a U-Net :) for binary segmentation. I will update the code when I have some spare time within the next month. However you can simply read this one and will soon notice the pattern after a bit
This repo includes Glioma Segmentation with Mask R-CNN and U-Net.