Unsupervised Change Detection Based on Image Reconstruction Loss

@article{Noh2022UnsupervisedCD,
  title={Unsupervised Change Detection Based on Image Reconstruction Loss},
  author={Hyeon-cheol Noh and Jin-gi Ju and Min-seok Seo and Jong-Dae Park and Dong-geol Choi},
  journal={2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)},
  year={2022},
  pages={1351-1360}
}
  • Hyeon-cheol NohJin-gi Ju Dong-geol Choi
  • Published 4 April 2022
  • Computer Science, Environmental Science
  • 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)
To train a change detector, bi-temporal images taken at different times in the same area are used. However, collecting labeled bi-temporal images is expensive and time consuming. To solve this problem, various unsupervised change detection methods have been proposed, but they still require unlabeled bi-temporal images. In this paper, we propose an unsupervised change detection method based on image reconstruction loss, which uses only a single-temporal unlabeled image. The image reconstruction… 

Figures and Tables from this paper

References

SHOWING 1-10 OF 34 REFERENCES

Unsupervised Change Detection in Satellite Images With Generative Adversarial Network

A novel change detection framework utilizing a special neural network architecture—Generative Adversarial Network (GAN) to generate many better coregistered images that are less sensitive to the problem of unregistered images and makes most of the deep learning structure.

A Spatial-Temporal Attention-Based Method and a New Dataset for Remote Sensing Image Change Detection

This work proposes a novel Siamese-based spatial–temporal attention neural network, which improves the F1-score of the baseline model from 83.9 to 87.3 with acceptable computational overhead and introduces a CD dataset LEVIR-CD, which is two orders of magnitude larger than other public datasets of this field.

Remote Sensing Image Change Detection With Transformers

This work proposes a bitemporal image transformer (BIT) to efficiently and effectively model contexts within the spatial-temporal domain and significantly outperforms the purely convolutional baseline using only three times lower computational costs and model parameters.

Unsupervised Deep Noise Modeling for Hyperspectral Image Change Detection

A novel noise modeling-based unsupervised fully convolutional network (FCN) framework is presented for HSI change detection, which utilizes the change detection maps of existing un supervised change detection methods to train the deep CNN, and then removes the noise during the end-to-end training process.

Unsupervised Change Detection in Satellite Images Using Convolutional Neural Networks

The proposed change detection method is unsupervised, and can be performed using any CNN model pre-trained for semantic segmentation, as well as to classify the detected changes into the correct semantic classes.

Object‐based change detection using correlation image analysis and image segmentation

This study introduces change detection based on object/neighbourhood correlation image analysis and image segmentation techniques and found that object‐based change classifications incorporating the OCIs or the NCIs produced more accurate change detection classes than other change detection results.

Multitask learning for large-scale semantic change detection

MVTec AD — A Comprehensive Real-World Dataset for Unsupervised Anomaly Detection

This work introduces the MVTec Anomaly Detection (MVTec AD) dataset containing 5354 high-resolution color images of different object and texture categories, and conducts a thorough evaluation of current state-of-the-art unsupervised anomaly detection methods based on deep architectures such as convolutional autoencoders, generative adversarial networks, and feature descriptors using pre-trained convolved neural networks.

Slow Feature Analysis for Change Detection in Multispectral Imagery

This paper proposes a novel slow feature analysis (SFA) algorithm for change detection that performs better in detecting changes than the other state-of-the-art change detection methods.

Constrained optical flow for aerial image change detection

A novel approach for change detection that considers the geometric aspects of camera sensors as well as the statistical properties of changes, based on optical flow matching, constrained by the epipolar geometry, and combined with a statistical change decision criterion is introduced.