114 citations · 276 across the 15 of their papers we have counts for
18 papers
LAMES: A Large-Scale and Artisanal Mining Environmental Segmentation Dataset
Matthias Kahl, Zhaiyu Chen, Sudipan Saha +3
Mining operations are of utmost importance to the economy of some nations. However, such operations result in land-use change, very high energy consumption, and negative impacts on…
Beyond Grid Data: Exploring Graph Neural Networks for Earth Observation
Shan Zhao, Zhaiyu Chen, Zhitong Xiong +3
Earth Observation (EO) data analysis has been significantly revolutionized by deep learning (DL), with applications typically limited to grid-like data structures. Graph Neural Net…
Segment Using Just One Example
Pratik Vora, Sudipan Saha
Semantic segmentation is an important topic in computer vision with many relevant application in Earth observation. While supervised methods exist, the constraints of limited annot…
Specialized Change Detection using Segment Anything
Tahir Ahmad, Sudipan Saha
Change detection (CD) is a fundamental task in Earth observation. While most change detection methods detect all changes, there is a growing need for specialized methods targeting…
Confidence Estimation in Unsupervised Deep Change Vector Analysis
Sudipan Saha
Unsupervised transfer learning-based change detection methods exploit the feature extraction capability of pre-trained networks to distinguish changed pixels from the unchanged one…
Exploring Geometric Deep Learning For Precipitation Nowcasting
Shan Zhao, Sudipan Saha, Zhitong Xiong +2
Precipitation nowcasting (up to a few hours) remains a challenge due to the highly complex local interactions that need to be captured accurately. Convolutional Neural Networks rel…
Deep Unsupervised Learning for 3D ALS Point Cloud Change Detection
Iris de Gélis, Sudipan Saha, Muhammad Shahzad +3
Change detection from traditional \added{2D} optical images has limited capability to model the changes in the height or shape of objects. Change detection using 3D point cloud \ad…
Spatial Context Awareness for Unsupervised Change Detection in Optical Satellite Images
Lukas Kondmann, Aysim Toker, Sudipan Saha +3
Detecting changes on the ground in multitemporal Earth observation data is one of the key problems in remote sensing. In this paper, we introduce Sibling Regression for Optical Cha…
Reiterative Domain Aware Multi-Target Adaptation
Sudipan Saha, Shan Zhao, Nasrullah Sheikh +1
Most domain adaptation methods focus on single-source-single-target adaptation settings. Multi-target domain adaptation is a powerful extension in which a single classifier is lear…
Segmentation of VHR EO Images using Unsupervised Learning
Sudipan Saha, Lichao Mou, Muhammad Shahzad +1
Semantic segmentation is a crucial step in many Earth observation tasks. Large quantity of pixel-level annotation is required to train deep networks for semantic segmentation. Eart…
Generative modeling of spatio-temporal weather patterns with extreme event conditioning
Konstantin Klemmer, Sudipan Saha, Matthias Kahl +2
Deep generative models are increasingly used to gain insights in the geospatial data domain, e.g., for climate data. However, most existing approaches work with temporal snapshots…
Trusting small training dataset for supervised change detection
Sudipan Saha, Biplab Banerjee, Xiao Xiang Zhu
Deep learning (DL) based supervised change detection (CD) models require large labeled training data. Due to the difficulty of collecting labeled multi-temporal data, unsupervised…
Out-of-distribution detection in satellite image classification
Jakob Gawlikowski, Sudipan Saha, Anna Kruspe +1
In satellite image analysis, distributional mismatch between the training and test data may arise due to several reasons, including unseen classes in the test data and differences…
Deep Reinforcement Learning for Band Selection in Hyperspectral Image Classification
Lichao Mou, Sudipan Saha, Yuansheng Hua +3
Band selection refers to the process of choosing the most relevant bands in a hyperspectral image. By selecting a limited number of optimal bands, we aim at speeding up model train…
Self-Supervised Multisensor Change Detection
Sudipan Saha, Patrick Ebel, Xiao Xiang Zhu
Most change detection methods assume that pre-change and post-change images are acquired by the same sensor. However, in many real-life scenarios, e.g., natural disaster, it is mor…
Towards Bridging the gap between Empirical and Certified Robustness against Adversarial Examples
Jay Nandy, Sudipan Saha, Wynne Hsu +2
The current state-of-the-art defense methods against adversarial examples typically focus on improving either empirical or certified robustness. Among them, adversarially trained (…
Ultrasound Image Classification using ACGAN with Small Training Dataset
Sudipan Saha, Nasrullah Sheikh
B-mode ultrasound imaging is a popular medical imaging technique. Like other image processing tasks, deep learning has been used for analysis of B-mode ultrasound images in the las…
Federated Transfer Learning: concept and applications
Sudipan Saha, Tahir Ahmad
Development of Artificial Intelligence (AI) is inherently tied to the development of data. However, in most industries data exists in form of isolated islands, with limited scope o…