258 Hits in 6.8 sec

Self-Supervised Pre-Training of Transformers for Satellite Image Time Series Classification

Yuan Yuan, Lei Lin
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Satellite image time series (SITS) classification is a major research topic in remote sensing and is relevant for a wide range of applications.  ...  The main idea of our proposal is to leverage the inherent temporal structure of satellite time series to learn general-purpose spectral-temporal representations related to land cover semantics.  ...  has advantages over CNNs and RNNs for satellite time series classification.  ... 
doi:10.1109/jstars.2020.3036602 fatcat:3dhovjahk5eupgkmkuj3ctig74

Multiscale Convolutional Transformer with Center Mask Pretraining for Hyperspectral Image Classification [article]

Sen Jia, Yifan Wang
2022 arXiv   pre-print
with Transformer network.In order to make more efficient use of unlabeled data, we propose a new self-supervised pretask.  ...  However, CNN-based methods are difficult to capture long-range dependencies, and also require a large amount of labeled data for model training.Besides, most of the self-supervised training methods in  ...  self-supervised learning process.  ... 
arXiv:2203.04771v4 fatcat:v4f4yv5xkzhudeffk4zuvrjkjy

Semi-Supervised Convolutional Long Short-Term Memory Neural Networks for Time Series Land Cover Classification

Jing Shen, Chao Tao, Ji Qi, Hao Wang
2021 Remote Sensing  
Time series images with temporal features are beneficial to improve the classification accuracy.  ...  Besides, it is a robust classification algorithm for time series optical images with cloud coverage, which reduces the requirements for cloudless remote sensing images and can be widely used in areas that  ...  The authors also sincerely thank the anonymous reviewers for their very competent comments and helpful suggestions. Conflicts of Interest: All authors declare no conflict of interest.  ... 
doi:10.3390/rs13173504 fatcat:nrifvolz5jcxbcrhnjzxzemkeq

Viewmaker Networks: Learning Views for Unsupervised Representation Learning [article]

Alex Tamkin, Mike Wu, Noah Goodman
2021 arXiv   pre-print
Many recent methods for unsupervised representation learning train models to be invariant to different "views," or distorted versions of an input.  ...  Remarkably, when pretraining on CIFAR-10, our learned views enable comparable transfer accuracy to the well-tuned SimCLR augmentations – despite not including transformations like cropping or color jitter  ...  However, compared to images, there is considerably less work on self-supervised learning and data augmentations for speech data.  ... 
arXiv:2010.07432v2 fatcat:xg3o7usfxbhppjb3afuiewnrle

Semi-MCNN: A Semi-supervised Multi-CNN Ensemble Learning Method for Urban Land Cover Classification Using Sub-meter HRRS Images

Runyu Fan, Ruyi Feng, Lizhe Wang, Jining Yan, Xiaohan Zhang
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Sub-meter high-resolution remote sensing (HRRS) image land cover classification could provide significant help for urban monitoring, management, and planning.  ...  This could significantly improve the generalization ability of the semi-supervised model, as well as the classification accuracy.  ...  ACKNOWLEDGMENT This paper is funded by National Natural Science Foundation of China (No. U1711266, No. 41925007 and No.41701429).  ... 
doi:10.1109/jstars.2020.3019410 fatcat:jtzkx4uhojh5jh6nmnrmgim3ce

Assessing Post-Disaster Damage from Satellite Imagery using Semi-Supervised Learning Techniques [article]

Jihyeon Lee, Joseph Z. Xu, Kihyuk Sohn, Wenhan Lu, David Berthelot, Izzeddin Gur, Pranav Khaitan, Ke-Wei Huang, Kyriacos Koupparis, Bernhard Kowatsch
2020 arXiv   pre-print
This paper shows a novel application of semi-supervised learning (SSL) to train models for damage assessment with a minimal amount of labeled data and large amount of unlabeled data.  ...  Recent research combines machine learning with remote sensing to automatically extract such information from satellite imagery, reducing manual labor and turn-around time.  ...  We support WFP innovators and external start-ups and companies through financial support, access to a network of experts and a global field reach.  ... 
arXiv:2011.14004v1 fatcat:beiddlsfynddbf5ad7ayqrsyuy

Class-agnostic Object Detection with Multi-modal Transformer [article]

Muhammad Maaz, Hanoona Rasheed, Salman Khan, Fahad Shahbaz Khan, Rao Muhammad Anwer, Ming-Hsuan Yang
2022 arXiv   pre-print
For the first time in literature, we demonstrate that Multi-modal Vision Transformers (MViT) trained with aligned image-text pairs can effectively bridge this gap.  ...  We show the significance of MViT proposals in a diverse range of applications including open-world object detection, salient and camouflage object detection, supervised and self-supervised detection tasks  ...  Self-supervised learning.  ... 
arXiv:2111.11430v3 fatcat:y2uw2c7ehjempantywt653edc4

Geographical Knowledge-driven Representation Learning for Remote Sensing Images [article]

Wenyuan Li, Keyan Chen, Hao Chen, Zhenwei Shi
2021 arXiv   pre-print
It contains 1,431,950 remote sensing images from Gaofen series satellites with various resolutions.  ...  The proliferation of remote sensing satellites has resulted in a massive amount of remote sensing images.  ...  They derive supervision information from the data itself via a series of pretext tasks. The design of pretext tasks is critical to the success of self-supervised learning.  ... 
arXiv:2107.05276v1 fatcat:wb3zzvn54jf4dj3blyvoonlvbi

TOV: The Original Vision Model for Optical Remote Sensing Image Understanding via Self-supervised Learning [article]

Chao Tao, Ji Qia, Guo Zhang, Qing Zhu, Weipeng Lu, Haifeng Li
2022 arXiv   pre-print
tasks, including scene classification, object detection, and semantic segmentation, and outperforms dominant ImageNet supervised pretrained method as well as two recently proposed SSL pretrained methods  ...  Do we on the right way for remote sensing image understanding (RSIU) by training models via supervised data-dependent and task-dependent way, instead of human vision in a label-free and task-independent  ...  A self-supervised pretraining method used to train TOV 4) MoCov2 [35, 42] . MoCov2 is also a self-supervised pretraining method used to train TOV model.  ... 
arXiv:2204.04716v1 fatcat:zd3cmazrdrdtfj5r2zph3hhbgq

Applying Self-Supervised Learning to Medicine: Review of the State of the Art and Medical Implementations

Alexander Chowdhury, Jacob Rosenthal, Jonathan Waring, Renato Umeton
2021 Informatics  
Self-supervised learning (SSL) is a growing area of machine learning that can take advantage of unlabeled data.  ...  Contrary to other machine learning paradigms, SSL algorithms create artificial supervisory signals from unlabeled data and pretrain algorithms on these signals.  ...  Acknowledgments: We would like to thank Jason Johnson and all members of the Artificial Intelligence Operations and Data Science Services group for their continuous support and the critical conversations  ... 
doi:10.3390/informatics8030059 fatcat:6osdf2ybknf37ojndtpszshxvy

Transformer-Based Decoder Designs for Semantic Segmentation on Remotely Sensed Images

Teerapong Panboonyuen, Kulsawasd Jitkajornwanich, Siam Lawawirojwong, Panu Srestasathiern, Peerapon Vateekul
2021 Remote Sensing  
First, utilizing the pre-training Swin Transformer (SwinTF) under Vision Transformer (ViT) as a backbone, the model weights downstream tasks by joining task layers upon the pretrained encoder.  ...  The results are compared with other image labeling state of the art (SOTA) methods, such as global convolutional network (GCN) and ViT.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/rs13245100 fatcat:rue3fsmvbrcedfzg7t7vdyn4cq

BreizhCrops: A Time Series Dataset for Crop Type Mapping [article]

Marc Rußwurm, Charlotte Pelletier, Maximilian Zollner, Sébastien Lefèvre, Marco Körner
2020 arXiv   pre-print
We present Breizhcrops, a novel benchmark dataset for the supervised classification of field crops from satellite time series.  ...  We aggregated label data and Sentinel-2 top-of-atmosphere as well as bottom-of-atmosphere time series in the region of Brittany (Breizh in local language), north-east France.  ...  For the temporal task of crop type mapping from satellite image time series, novel approaches have predominantly been tested on self-created datasets and only partly compared to other state-of-the-art  ... 
arXiv:1905.11893v2 fatcat:nyz67pium5b3tczoxcgekd7gdm

Semantic Segmentation of Remote Sensing Images with Self-supervised Multi-task Representation Learning

Wenyuan Li, Hao Chen, Zhenwei Shi
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
The above self-supervised methods designed for natural images do not consider the characteristics of remote sensing images and may not work properly.  ...  affects the performance of semantic segmentation for remote sensing images.  ...  Self-supervised Representation Learning for Remote Sensing Images Although researches of self-supervised representation learning for natural images are developing rapidly, methods for remote sensing images  ... 
doi:10.1109/jstars.2021.3090418 fatcat:e7c63tbe5rfvzc5bhn5cc6zt4m

Remote Sensing Images Semantic Segmentation with General Remote Sensing Vision Model via a Self-Supervised Contrastive Learning Method [article]

Haifeng Li, Yi Li, Guo Zhang, Ruoyun Liu, Haozhe Huang, Qing Zhu, Chao Tao
2021 arXiv   pre-print
However, most of the existing contrastive learning is designed for classification tasks to obtain an image-level representation, which may be sub-optimal for semantic segmentation tasks requiring pixel-level  ...  Our study promotes the development of self-supervised learning in the field of remote sensing semantic segmentation. The source code is available at  ...  Such a situation is available, as we can easily obtain a large number of images from the same source through satellite technology. 2) Effect of the amount of self-supervised data: Since the self-supervised  ... 
arXiv:2106.10605v1 fatcat:eohfunmdrzdflmhkupdda4wyry

Deep Learning in Remote Sensing: A Comprehensive Review and List of Resources

Xiao Xiang Zhu, Devis Tuia, Lichao Mou, Gui-Song Xia, Liangpei Zhang, Feng Xu, Friedrich Fraundorfer
2017 IEEE Geoscience and Remote Sensing Magazine  
In this article, we analyze the challenges of using deep learning for remote sensing data analysis, review the recent advances, and provide resources to make deep learning in remote sensing ridiculously  ...  This capability is triggering a shift from individual image analysis to time-series processing.  ...  . ◗ More recent work addresses the transferability of deep learning for change detection, while analyzing data of long time series for large-scale problems.  ... 
doi:10.1109/mgrs.2017.2762307 fatcat:ec7b32lpdnhvzbdz2uoayw6anq
« Previous Showing results 1 — 15 out of 258 results