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

  title={TOV: The Original Vision Model for Optical Remote Sensing Image Understanding via Self-supervised Learning},
  author={Chao Tao and Jirong Qia and Guo Zhang and Qing Zhu and Weipeng Lu and Haifeng Li},
Do we on the right way for remote sensing image understanding (RSIU) by training models via supervised data-dependent and task-dependent way, in-stead of human vision in a label-free and task-independent way? We argue that a more desirable RSIU model should be trained with intrinsic structure from data rather that extrinsic human labels to realize generalizability across a wide range of RSIU tasks. According to this hypothesis, we proposed T he O riginal V ision model (TOV) in remote sensing… 

Consecutive Pretraining: A Knowledge Transfer Learning Strategy with Relevant Unlabeled Data for Remote Sensing Domain

By utilizing the proposed CSPT for task-aware model training, almost all downstream tasks in the RSD can outperform the previous knowledge transfer learning strategies based on model pre-training without any expensive manually labeling and even surpass the state-of-the-art (SOTA) performance without any careful network architecture designing.

Self-Supervised Learning for Scene Classification in Remote Sensing: Current State of the Art and Perspectives

This article reviews the underlying principles developed by various self-supervised methods with a focus on scene classification task, and investigates the impact of individual augmentations when applied to remote sensing data as well as the use of self- supervised pre-training to boost the classification performance with limited number of labeled samples.

Self-supervised remote sensing feature learning: Learning Paradigms, Challenges, and Future Works

This paper analyzes SSFL over the other two learning paradigms in RSIs understanding tasks and gives a comprehensive review of the existing SSFL work in RS, including the pre-training dataset, self- supervised feature learning signals, and the evaluation methods.

HyperNet: Self-Supervised Hyperspectral Spatial–Spectral Feature Understanding Network for Hyperspectral Change Detection

The fast development of self-supervised learning (SSL) lowers the bar learning feature representation from massive unlabeled data and has triggered a series of researches on change detection of

Semantic-Aware Dense Representation Learning for Remote Sensing Image Change Detection

Different from traditional supervised pretraining that learns the mapping from image to label, semantic supervision is incorporated into the self-supervised learning (SSL) framework and significantly outperforms ImageNet pretraining, in-domain supervision, and several SSL methods.



Remote Sensing Image Scene Classification With Self-Supervised Paradigm Under Limited Labeled Samples

This work introduces a new self-supervised learning (SSL) mechanism to obtain the high-performance pretraining model for RSI scene classification from large unlabeled data and demonstrates that this new learning paradigm outperforms the traditional dominant ImageNet pretrained model.

Global and Local Contrastive Self-Supervised Learning for Semantic Segmentation of HR Remote Sensing Images

A global style and local matching contrastive learning network (GLCNet) for RSI semantic segmentation and shows that this method mostly outperforms the state-of-the-art self-supervised methods and the ImageNet pretraining method when there are some differences between the datasets of upstream tasks and downstream tasks.

PatternNet: A Benchmark Dataset for Performance Evaluation of Remote Sensing Image Retrieval

DeepGlobe 2018: A Challenge to Parse the Earth through Satellite Images

  • Ilke DemirK. Koperski R. Raskar
  • Computer Science
    2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
  • 2018
The DeepGlobe 2018 Satellite Image Understanding Challenge is presented, which includes three public competitions for segmentation, detection, and classification tasks on satellite images, and characteristics of each dataset are analyzed, and evaluation criteria for each task are defined.

Performance Evaluation of Single-Label and Multi-Label Remote Sensing Image Retrieval Using a Dense Labeling Dataset

DLRSD provided the literature a benchmark for RSIR and other pixel-based problems such as semantic segmentation and demonstrated the advantages of multiple labels over single labels for interpreting complex remote sensing images.

Self-supervised Pretraining of Visual Features in the Wild

The final SElf-supERvised (SEER) model, a RegNetY with 1.3B parameters trained on 1B random images with 512 GPUs achieves 84.2% top-1 accuracy, surpassing the best self-supervised pretrained model by 1% and confirming that self- Supervised learning works in a real world setting.

RSI-CB: A Large-Scale Remote Sensing Image Classification Benchmark Using Crowdsourced Data

The experiments show that RSI-CB is more suitable as a benchmark for remote sensing image classification tasks than other benchmarks in the big data era and has many potential applications.

Self-Supervised Visual Feature Learning With Deep Neural Networks: A Survey

An extensive review of deep learning-based self-supervised general visual feature learning methods from images or videos as a subset of unsupervised learning methods to learn general image and video features from large-scale unlabeled data without using any human-annotated labels is provided.

Random Access Memories: A New Paradigm for Target Detection in High Resolution Aerial Remote Sensing Images

Experimental results on a public remote sensing target detection data set show the proposed new paradigm “random access memories (RAM)” outperforms several other state of the art methods.