The field of remote sensing and GeoAI is rapidly advancing with the development of new deep learning models and techniques. One of the key directions is the use of multimodal perception systems that can integrate complementary information from different sensors, such as RGB, depth, and thermal imagery. This enables safe navigation in unstructured environments and accurate segmentation of complex terrain. Another area of focus is the development of unsupervised and weakly supervised learning methods that can reduce the need for labeled data and enable the detection of objects and features in remote sensing images. The use of transformer-based neural networks and convolutional neural networks is also becoming increasingly popular in remote sensing applications, including object detection, image segmentation, and change detection. Noteworthy papers in this area include the OmniUnet model, which achieves state-of-the-art performance in semantic segmentation using RGB, depth, and thermal imagery, and the IAMAP plugin, which enables non-AI specialists to leverage deep learning methods for remote sensing image analysis. The TNet model is also notable for its ability to progressively integrate low-resolution features into higher-resolution features, enabling spatially-aware convolutional kernels that blend global and local information.
Advancements in Remote Sensing and GeoAI
Sources
Leveraging Convolutional and Graph Networks for an Unsupervised Remote Sensing Labelling Tool
OmniUnet: A Multimodal Network for Unstructured Terrain Segmentation on Planetary Rovers Using RGB, Depth, and Thermal Imagery
IAMAP: Unlocking Deep Learning in QGIS for non-coders and limited computing resources
Evaluation and Analysis of Deep Neural Transformers and Convolutional Neural Networks on Modern Remote Sensing Datasets
Infrared Object Detection with Ultra Small ConvNets: Is ImageNet Pretraining Still Useful?
Tobler's First Law in GeoAI: A Spatially Explicit Deep Learning Model for Terrain Feature Detection Under Weak Supervision
Modular Transformer Architecture for Precision Agriculture Imaging
Deep learning framework for crater detection and identification on the Moon and Mars
Investigating the Impact of Large-Scale Pre-training on Nutritional Content Estimation from 2D Images
Prototype-Driven Structure Synergy Network for Remote Sensing Images Segmentation
TNet: Terrace Convolutional Decoder Network for Remote Sensing Image Semantic Segmentation
Deeper Inside Deep ViT
RPCANet++: Deep Interpretable Robust PCA for Sparse Object Segmentation
A machine learning approach for image classification in synthetic aperture RADAR
Wavelet-Guided Dual-Frequency Encoding for Remote Sensing Change Detection
CoCAViT: Compact Vision Transformer with Robust Global Coordination