Climate change has led to an increased frequency of natural disasters such as floods and cyclones.
The Detection Transformer (DETR) has emerged as a pivotal role in object detection tasks, setting new performance benchmarks due to its end-to-end design and scalability.
Approaches for appraising feature importance approximations, alternatively referred to as attribution methods, have been established across an extensive array of contexts.
Recently, research in the remote sensing field has focused primarily on the pretraining method and the size of the dataset, with limited emphasis on the number of model parameters.
Ranked #1 on Object Detection In Aerial Images on DIOR-R
Future frame prediction has been approached through two primary methods: autoregressive and non-autoregressive.
Ranked #1 on Weather Forecasting on SEVIR
For change detection in remote sensing, constructing a training dataset for deep learning models is difficult due to the requirements of bi-temporal supervision.
Adversarial examples, crafted by adding imperceptible perturbations to natural inputs, can easily fool deep neural networks (DNNs).
In the training of deep learning models, how the model parameters are initialized greatly affects the model performance, sample efficiency, and convergence speed.
In deep learning-based object detection on remote sensing domain, nuisance factors, which affect observed variables while not affecting predictor variables, often matters because they cause domain changes.
In this paper, we revisit the classical bootstrap aggregating approach in the context of modern transfer learning for data-efficient disaster damage detection.
Road extraction from very high resolution satellite (VHR) images is one of the most important topics in the field of remote sensing.
Adversarial training is a training scheme designed to counter adversarial attacks by augmenting the training dataset with adversarial examples.
Saliency Map, the gradient of the score function with respect to the input, is the most basic technique for interpreting deep neural network decisions.
Object detection and classification for aircraft are the most important tasks in the satellite image analysis.
SmoothGrad and VarGrad are techniques that enhance the empirical quality of standard saliency maps by adding noise to input.