Self-Supervised Global–Local Contrastive Learning for Fine-Grained Change Detection in VHR Images

Jiang, Fenlong; Gong, Maoguo; Zheng, Hanhong; Liu, Tongfei; Zhang, Mingyang; Liu, Jialu

doi:10.1109/tgrs.2023.3238327

Cited by 20 publications

(7 citation statements)

References 61 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…While these methods demonstrate the potential application of self-supervised methods in CD tasks, their performance lags behind fully supervised methods due to their unsupervised nature. On the other hand, self-supervised methods for high-resolution image CD have been proposed, as in references [201][202][203][204], but they still rely on a substantial amount of supervised data during fine-tuning, not fully addressing the challenges of data annotation. To date, only reference [205] has experimented with fine-tuning self-supervised models on a minimal dataset (1%), but there remains significant room for performance improvement.…”

Section: Self-supervised Learningmentioning

confidence: 99%

Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms

Wang,

Zhang,

Gao

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

Change detection (CD) in remote sensing (RS) imagery is a pivotal method for detecting changes in the Earth’s surface, finding wide applications in urban planning, disaster management, and national security. Recently, deep learning (DL) has experienced explosive growth and, with its superior capabilities in feature learning and pattern recognition, it has introduced innovative approaches to CD. This review explores the latest techniques, applications, and challenges in DL-based CD, examining them through the lens of various learning paradigms, including fully supervised, semi-supervised, weakly supervised, and unsupervised. Initially, the review introduces the basic network architectures for CD methods using DL. Then, it provides a comprehensive analysis of CD methods under different learning paradigms, summarizing commonly used frameworks. Additionally, an overview of publicly available datasets for CD is offered. Finally, the review addresses the opportunities and challenges in the field, including: (a) incomplete supervised CD, encompassing semi-supervised and weakly supervised methods, which is still in its infancy and requires further in-depth investigation; (b) the potential of self-supervised learning, offering significant opportunities for Few-shot and One-shot Learning of CD; (c) the development of Foundation Models, with their multi-task adaptability, providing new perspectives and tools for CD; and (d) the expansion of data sources, presenting both opportunities and challenges for multimodal CD. These areas suggest promising directions for future research in CD. In conclusion, this review aims to assist researchers in gaining a comprehensive understanding of the CD field.

show abstract

Section: Self-supervised Learningmentioning

confidence: 99%

Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms

Wang,

Zhang,

Gao

et al. 2024

Remote Sensing

View full text Add to dashboard Cite

show abstract

“…Similarly, Ramkumar et al [60,61] proposed a self-supervised pre-training method for natural image scene CD tasks. Jiang et al [62] proposed a self-supervised global-local contrastive learning (GLCL) framework that extends instance discrimination to pixel-level CD tasks. Through GLCL, features from the same instance with different views are pulled closer together while features from different instances are separated, enhancing the discriminative feature representation from both global and local perspectives for downstream CD tasks.…”

Section: Use Of Ssl In Remote Sensing CDmentioning

confidence: 99%

“…The approach taken in this study was inspired by the methodologies of previous studies [62][63][64]. It leverages bi-temporal remote sensing images of the study area as positive sample pairs for SSL, effectively eliminating the need for additional data collection during the pre-training phase.…”

Section: Use Of Ssl In Remote Sensing CDmentioning

confidence: 99%

Detection of Changes in Buildings in Remote Sensing Images via Self-Supervised Contrastive Pre-Training and Historical Geographic Information System Vector Maps

Feng,

Guan,

et al. 2023

Remote Sensing

View full text Add to dashboard Cite

The detection of building changes (hereafter ‘building change detection’, BCD) is a critical issue in remote sensing analysis. Accurate BCD faces challenges, such as complex scenes, radiometric differences between bi-temporal images, and a shortage of labelled samples. Traditional supervised deep learning requires abundant labelled data, which is expensive to obtain for BCD. By contrast, there is ample unlabelled remote sensing imagery available. Self-supervised learning (SSL) offers a solution, allowing learning from unlabelled data without explicit labels. Inspired by SSL, we employed the SimSiam algorithm to acquire domain-specific knowledge from remote sensing data. Then, these well-initialised weight parameters were transferred to BCD tasks, achieving optimal accuracy. A novel framework for BCD was developed using self-supervised contrastive pre-training and historical geographic information system (GIS) vector maps (HGVMs). We introduced the improved MS-ResUNet network for the extraction of buildings from new temporal satellite images, incorporating multi-scale pyramid image inputs and multi-layer attention modules. In addition, we pioneered a novel spatial analysis rule for detecting changes in building vectors in bi-temporal images. This rule enabled automatic BCD by harnessing domain knowledge from HGVMs and building upon the spatial analysis of building vectors in bi-temporal images. We applied this method to two extensive datasets in Liuzhou, China, to assess its effectiveness in both urban and suburban areas. The experimental results demonstrated that our proposed approach offers a competitive quantitative and qualitative performance, surpassing existing state-of-the-art methods. Combining HGVMs and high-resolution remote sensing imagery from the corresponding years is useful for building updates.

show abstract

“…Wang et al [8] introduce an extra backbone to learn fine-grained classification and adopt knowledge distillation to keep it lightweight. Ouyang et al [65] propose PCLDet to maximize the interclass distance and minimize the intraclass distance by prototypical contrastive learning. Cheng et al [9] propose SFRNet with a spatial and channel transformer to capture discriminative features and adopt metric learning to enhance the separability of fine-grained classes.…”

Section: Fine-grained Object Detectionmentioning

confidence: 99%

HTD: Heterogeneous Task Decoupling for Two-Stage Object Detection

Chen

Li³

et al. 2021

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Fine-grained object detection (FGOD) extends object detection with the capability of fine-grained recognition. In recent two-stage FGOD methods, the region proposal serves as a crucial link between detection and fine-grained recognition. However, current methods overlook that some proposalrelated procedures inherited from general detection are not equally suitable for FGOD, limiting the multi-task learning from generation, representation, to utilization. In this paper, we present PETDet (Proposal Enhancement for Two-stage finegrained object detection) to better handle the sub-tasks in twostage FGOD methods. Firstly, an anchor-free Quality Oriented Proposal Network (QOPN) is proposed with dynamic label assignment and attention-based decomposition to generate highquality oriented proposals. Additionally, we present a Bilinear Channel Fusion Network (BCFN) to extract independent and discriminative features of the proposals. Furthermore, we design a novel Adaptive Recognition Loss (ARL) which offers guidance for the R-CNN head to focus on high-quality proposals. Extensive experiments validate the effectiveness of PETDet. Quantitative analysis reveals that PETDet with ResNet50 reaches state-of-theart performance on various FGOD datasets, including FAIR1M-v1.0 (42.96 AP), FAIR1M-v2.0 (48.81 AP), MAR20 (85.91 AP) and ShipRSImageNet (74.90 AP). The proposed method also achieves superior compatibility between accuracy and inference speed. Our code and models will be released at https://github.com/ canoe-Z/PETDet.

show abstract

Self-Supervised Global–Local Contrastive Learning for Fine-Grained Change Detection in VHR Images

Cited by 20 publications

References 61 publications

Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms

Advances and Challenges in Deep Learning-Based Change Detection for Remote Sensing Images: A Review through Various Learning Paradigms

Detection of Changes in Buildings in Remote Sensing Images via Self-Supervised Contrastive Pre-Training and Historical Geographic Information System Vector Maps

HTD: Heterogeneous Task Decoupling for Two-Stage Object Detection

Contact Info

Product

Resources

About