Filters








247 Hits in 3.4 sec

On Auxiliary Losses for Semi-Supervised Semantic Segmentation

Javiera Castillo-Navarro, Bertrand Le Saux, Alexandre Boulch, Sébastien Lefèvre
2020 European Conference on Principles of Data Mining and Knowledge Discovery  
Indeed, labeled remote sensing data are scarce and likely insufficient to train fully supervised models with good generalization capacities.  ...  This work addresses the problem of semisupervised semantic segmentation from a multi-task learning perspective.  ...  Application of self-supervised methods in remote sensing is very recent.  ... 
dblp:conf/pkdd/Castillo-Navarro20 fatcat:3xu2zy7ksrc57precr6xigaijq

EVALUATION OF SEMI-SUPERVISED LEARNING FOR CNN-BASED CHANGE DETECTION

E. Bousias Alexakis, C. Armenakis
2021 The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences  
In this work we experiment with the implementation of a semi-supervised training approach in an attempt to improve the image semantic segmentation performance of models trained using a small number of  ...  The approach is based on the Mean Teacher method, a semi-supervised approach, successfully applied for image classification and for sematic segmentation of medical images.  ...  ACKNOWLEDGEMENTS This work is financially supported by the Natural Sciences and Engineering Research Council of Canada (NSERC Discovery and CREATE grants) and York University.  ... 
doi:10.5194/isprs-archives-xliii-b3-2021-829-2021 fatcat:de3tnxo6sndlzayzc2zi5kctme

Research Progress on Few-Shot Learning for Remote Sensing Image Interpretation

Xian Sun, Bing Wang, Zhirui Wang, Hao Li, Hengchao Li, Kun Fu
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Then, three typical remote sensing interpretation applications are listed, including scene classification, semantic segmentation, and object detection, together with the corresponding public datasets and  ...  The rapid development of deep learning brings effective solutions for remote sensing image interpretation.  ...  Typical Applications of Remote Sensing Image Interpretation 1) Scene Classification: Scene classification, which aims to automatically assign remote sensing images with the predefined semantic labels,  ... 
doi:10.1109/jstars.2021.3052869 fatcat:ldos3sx6mvaapjkgsua73l7tve

$\mathrm{BAS}^4$Net: Boundary-Aware Semi-Supervised Semantic Segmentation Network for Very High Resolution Remote Sensing Images

Xian Sun, Aijun Shi, Hai Huang, Helmut Mayer
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Semantic segmentation is a fundamental task in remote sensing image understanding.  ...  However, it is still challenging for Very High Resolution (VHR) remote sensing images.  ...  Citation information: DOI 10.1109/JSTARS.2020.3021098, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  ... 
doi:10.1109/jstars.2020.3021098 fatcat:unvmqs3he5fxzfxpjoxn44gjxi

SemiSANet: A Semi-Supervised High-Resolution Remote Sensing Image Change Detection Model Using Siamese Networks with Graph Attention

Chengzhe Sun, Jiangjiang Wu, Hao Chen, Chun Du
2022 Remote Sensing  
The model is trained to make the CD results of the distorted images consistent with the corresponding pseudo-label. Extensive experiments are conducted on two high-resolution remote sensing datasets.  ...  Change detection (CD) is one of the important applications of remote sensing and plays an important role in disaster assessment, land use detection, and urban sprawl tracking.  ...  In contrast, remote sensing images have much richer semantic information.  ... 
doi:10.3390/rs14122801 fatcat:4b44tresqbhhdnqot4rhtxihii

Cycle and Self-Supervised Consistency Training for Adapting Semantic Segmentation of Aerial Images

Han Gao, Yang Zhao, Peng Guo, Zihao Sun, Xiuwan Chen, Yunwei Tang
2022 Remote Sensing  
Semantic segmentation is a critical problem for many remote sensing (RS) image applications.  ...  The translated source data then drive a pipeline of supervised semantic segmentation model training.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/rs14071527 doaj:3e371156360c450ea7b23f242d99e34f fatcat:offdwang4va2hbolpbcwu2mp44

Patch-Free Bilateral Network for Hyperspectral Image Classification Using Limited Samples

Bing Liu, Xuchu Yu
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Jia for providing the open source codes of DLRGF and S-DMM, respectively.  ...  Nevertheless, HSI classification task is very different from semantic segmentation task. The training samples of semantic segmentation task are a group of labeled images.  ...  As for DL-based method, semisupervised learning methods such as self-training [34] and cotraining [35] have been applied to the training process of deep model.  ... 
doi:10.1109/jstars.2021.3121334 fatcat:te2app3kdza45kbmvh4qn3lr3y

Semi-Supervised Adversarial Semantic Segmentation Network Using Transformer and Multiscale Convolution for High-Resolution Remote Sensing Imagery

Yalan Zheng, Mengyuan Yang, Min Wang, Xiaojun Qian, Rui Yang, Xin Zhang, Wen Dong
2022 Remote Sensing  
Semantic segmentation is a crucial approach for remote sensing interpretation.  ...  In this study, a novel semi-supervised adversarial semantic segmentation network is developed for remote sensing information extraction.  ...  Conflicts of Interest: The authors declare no conflict of interest.  ... 
doi:10.3390/rs14081786 fatcat:dh3ilaolr5cp7i45p66rhr5k4u

Unsupervised Single-Scene Semantic Segmentation for Earth Observation

Sudipan Saha, Muhammad Shahzad, Lichao Mou, Qian Song, Xiao Xiang Zhu
2022 IEEE Transactions on Geoscience and Remote Sensing  
Generally, a large number of pixelwise labeled images are required to train deep models for supervised semantic segmentation.  ...  An important step in many Earth observation tasks is semantic segmentation.  ...  INTRODUCTION R APID development of remote sensing technologies has drastically increased the quantity of Earth observation sensors acquiring images with different spatial, spectral, and temporal resolution  ... 
doi:10.1109/tgrs.2022.3174651 fatcat:43tfdtkqvngx7dbd7rqpa27au4

Consistency-Regularized Region-Growing Network for Semantic Segmentation of Urban Scenes with Point-Level Annotations [article]

Yonghao Xu, Pedram Ghamisi
2022 arXiv   pre-print
To reduce the annotation burden, this paper proposes a consistency-regularized region-growing network (CRGNet) to achieve semantic segmentation of VHR images with point-level annotations.  ...  Deep learning algorithms have obtained great success in semantic segmentation of very high-resolution (VHR) images.  ...  The Vaihingen dataset was provided by the German Society for Photogrammetry, Remote Sensing and Geoinformation (DGPF).  ... 
arXiv:2202.03740v2 fatcat:un7c5txmgbcdvp3epvcofwbhpu

Evaluating Self and Semi-Supervised Methods for Remote Sensing Segmentation Tasks [article]

Chaitanya Patel, Shashank Sharma, Valerie J. Pasquarella, Varun Gulshan
2022 arXiv   pre-print
We quantify performance improvements on these remote sensing segmentation tasks when additional imagery outside of the original supervised dataset is made available for training.  ...  We perform a rigorous evaluation of SimCLR, a self-supervised method, and FixMatch, a semi-supervised method, on three remote sensing tasks: riverbed segmentation, land cover mapping, and flood mapping  ...  reviewing our experiments, Aparna Taneja for the help in generating the Sentinel-1 unlabeled data, and Oliver Guinan, Noel Gorelick, Christopher Brown, Wanda Czerwinski, and John Platt for reading drafts of  ... 
arXiv:2111.10079v2 fatcat:gzs3mu27wzfmze7octsl7ps2qy

Improved Metric Learning with CNN for Very High-Resolution Remote Sensing Image Classification

Cheng Shi, Lv ZhiYong, Huifang Shen, Li Fang, Zhenzhen You
2020 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Faced with the limited labeled samples on a high-resolution remote sensing image, a semisupervised method becomes an effective way.  ...  The number of labeled samples has a great impact on the classification results of a very-high-resolution (VHR) remote sensing image.  ...  The original VHR remote sensing image is segmented with the entropy rate segmentation (ERS) method [45] .  ... 
doi:10.1109/jstars.2020.3033944 fatcat:f7zdbuy3ibfgrdwuke7gpkkqyq

Semi-Supervised Semantic Segmentation in Earth Observation: The MiniFrance Suite, Dataset Analysis and Multi-task Network Study [article]

Javiera Castillo-Navarro, Bertrand Le Saux, Alexandre Boulch, Nicolas Audebert, Sébastien Lefèvre
2020 arXiv   pre-print
We introduce a novel large-scale dataset for semi-supervised semantic segmentation in Earth Observation, the MiniFrance suite.  ...  Nevertheless, the most distinctive quality of MiniFrance is being the only dataset in the field especially designed for semi-supervised learning: it contains labeled and unlabeled images in its training  ...  Semantic Segmentation Semantic segmentation consists in the process of assigning a class label to every pixel on an image.  ... 
arXiv:2010.07830v1 fatcat:h7k5dnh5nnhffl2iy4o6yg67du

Low-Shot Learning for the Semantic Segmentation of Remote Sensing Imagery

Ronald Kemker, Ryan Luu, Christopher Kanan
2018 IEEE Transactions on Geoscience and Remote Sensing  
vision are not ideal for remote sensing problems.  ...  "Self-taught" feature learning frameworks are trained with large quantities of unlabeled imagery, and then these networks extract spatial-spectral features from annotated data for supervised classification  ...  Michael Gartley with DIRSIG support. Acknowledgements This chapter is a reprint of the material as it appears in IEEE Transactions on Geoscience and Remote Sensing, Issue 99, 2018.  ... 
doi:10.1109/tgrs.2018.2833808 fatcat:svytkstfsnfnxhoigwumtifarm

Land-use Mapping for High Spatial Resolution Remote Sensing Image via Deep Learning: A Review

Ning Zang, Yun Cao, Yuebin Wang, Bo Huang, Liqiang Zhang, P Takis Mathiopoulos
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
Land-use mapping (LUM) using high spatial resolution remote sensing images (HSR-RSIs) is a challenging and crucial technology.  ...  We then briefly review the fundamentals and the developments of the development of semantic segmentation and single object segmentation.  ...  [203] proposed an FCN that consists of ResNet-34 and the decoder to automatic extract road. Nicolas et al. [204] applied SegNet to vehicle detection and segmentation in remote sensing images.  ... 
doi:10.1109/jstars.2021.3078631 fatcat:ubblvynhdvd2ndzhjablus22cy
« Previous Showing results 1 — 15 out of 247 results