Welcome to the repository for our research paper titled "Data Fusion and Models Integration for Enhanced Semantic Segmentation in Remote Sensing". This repository will be regularly updated following the publication of our paper to include the following key resources:
-
Data Fusion Method for Remote Sensing Semantic Segmentation. We will provide detailed methodologies and code implementations on various data fusion techniques specifically tailored for enhancing semantic segmentation in remote sensing imagery. These methods are designed to leverage the complementary information from multiple data sources to improve segmentation accuracy.
-
Training and Integration Tools for Remote Sensing Semantic Segmentation Models. This part will include scripts and guidelines for training semantic segmentation models on remote sensing data. We will also share our approaches for integrating multiple models to achieve higher performance through ensemble techniques.
-
GGDS Dataset. We will release the GGDS Dataset, a novel dataset created by fusing data from Google Maps and GlobeLand30. This dataset is specifically curated for training and evaluating semantic segmentation models in the context of remote sensing. Detailed descriptions, download links, and usage instructions will be provided.
Stay tuned for updates!