Development of an autonomous, task-distributed drone network based on ROS to improve the characterization of isolated, remote targets in a time and resource efficient manner.

A command line toolkit to generate maps, point clouds, 3D models and DEMs from drone, balloon or kite images.

Aerial Photographs of Colorado collection metadata from the University of Colorado Boulder. Aerial photographs dramatically portray the changing landscape of Colorado: a mountain valley can be seen where there is now a reservoir; changes in the vegetation and ground cover can be traced over the years, and the growth of towns and cities documente…

Web-Interface-GCP-Finder é um projeto que pretende simplificar a utilização da ferramenta GCP Finder, que em sí é um programa capaz de identificar Aruco Markers em imagens.

Native (unofficial) WebApp for Google Maps, built with Tauri

Zero shot image classification

This repository provides an implementation of semantic segmentation for road networks using PyTorch and the U-Net architecture. It focuses specifically on processing aerial images from the Massachusetts dataset.

Computer Vision AI to perform object detection in aerial images taken from drones and small aircraft. Flask API to run inference in the cloud.

This collection of MATLAB® functions is for working with Airborne Topographic Mapper (ATM) laser altimetry data products in HDF5 waveform format.

Coding used to process drone-captured Near-Infrared Images into Normalised Differential Vegetation Index (NDVI) greyscale images which are then further processed using both a segmentation of ndvi around the tomb region followed by a contour overlay in the perimeter of the tombs. Version 2 uses a standard contour, Version 4 is an attempt, with li…

Modified source code of published article: Real-ESRGAN: A deep learning approach for general image restoration and its application to aerial images

Cambridge air photos website

Adversarial Large-scale Root Gaps Inpainting

Load, Evaluate, Split, and Merge DOTA dataset

A tutorial providing an end-to-end workflow of image segmentation of buildings based on aerial images.

Deep Learning based Aerial Image Segmentation

Road network graph refinement with GNNs.

[This project was completed in September 2020] The GML-Net is a convolutional neural network (CNN) that is based on U-Net architecture with an encoder derived from the ResNet family and BottleNeck blocks that provide reading and aggregation of feature maps from a cross-section of various scales. Effective network learning is ensured by loss func…

Additional results for "Using Drones as Reference Sensors for Neural-Networks-Based Modeling of Automotive Perception Errors"

Convenient ways of downloading, tiling and finally converting aerial imagery of Berlin