CSANet: Cross-Temporal Interaction Symmetric Attention Network for Hyperspectral Image Change Detection

Song, Ruoxi; Ni, Weihan; Wei, Cheng; Wang, Xianghai

doi:10.1109/lgrs.2022.3179134

Cited by 41 publications

(15 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CFIN elegantly integrates CNN and Transformer, achieving competitive performance and a good balance between computational cost and performance. Different from the previous work [67], our CGA sent Q and K generated by the first CGA to the second CGA, and then add them with the new Q and K generated in the second CGA. After that, the result of the dot product of newly generated K and Q are dot products with V to obtain the final results.…”

Section: Discussionmentioning

confidence: 99%

Cross-receptive Focused Inference Network for Lightweight Image Super-Resolution

Li¹,

Li²,

Gao³

et al. 2022

Preprint

View full text Add to dashboard Cite

With the development of deep learning, single image super-resolution (SISR) has achieved significant breakthroughs. Recently, methods to enhance the performance of SISR networks based on global feature interactions have been proposed. However, the capabilities of neurons that need to adjust their function in response to the context dynamically are neglected. To address this issue, we propose a lightweight Cross-receptive Focused Inference Network (CFIN), a hybrid network composed of a Convolutional Neural Network (CNN) and a Transformer. Specifically, a novel Cross-receptive Field Guide Transformer (CFGT) is designed to adaptively modify the network weights by using modulated convolution kernels combined with local representative semantic information. In addition, a CNN-based Cross-scale Information Aggregation Module (CIAM) is proposed to make the model better focused on potentially practical information and improve the efficiency of the Transformer stage. Extensive experiments show that our proposed CFIN is a lightweight and efficient SISR model, which can achieve a good balance between computational cost and model performance.

show abstract

Section: Discussionmentioning

confidence: 99%

Cross-receptive Focused Inference Network for Lightweight Image Super-Resolution

Li¹,

Li²,

Gao³

et al. 2022

Preprint

View full text Add to dashboard Cite

show abstract

“…3) Comparison Methods: Eight competitive algorithms are opted for comparison and verify the effectiveness of GlobalMind, involving CVA [62], Iterative Slow Feature Analysis (ISFA) [63], GETNET [34], ML-EDAN [41], Bitemporal Image Transformer (BIT) [64], CSA-Net [54], EMS-Net [65], and SST-Former [53], covering most of the state-of-the-art CNN-, RNN-, transformer-based, and hybrid architecture-based approaches. GETNET (without unmixing) [34] is an end-to-end convolutional neural network, incorporating a novel difference affinity of bi-temporal HSIs to provide more abundant crosschannel gradient information.…”

Section: B Implementation Details 1) Experimenta L Set Tingsmentioning

confidence: 99%

“…CSA-Net [54] put forward e an original cross-temporal interaction symmetric attention network to dig the difference features oriented from each temporal feature embedding.…”

Section: B Implementation Details 1) Experimenta L Set Tingsmentioning

confidence: 99%

“…SST-Former [53] proposed a joint spectral, spatial, and temporal transformer for hyperspectral image change detection, where single spectral curves are encoded to extract spectral sequence information and the spatial transformer encoder is used to extract spatial texture information from local patches. CSA-Net [54] put forward a cross-temporal interaction symmetric attention module to integrate the difference features oriented from each temporal feature embedding. CDFormer [55] change detection map.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-Temporal Spatial-Spectral Comparison Network For Hyperspectral Anomalous Change Detection

2022

IGARSS 2022 - 2022 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

High spectral resolution imagery of the Earth's surface enables users to monitor changes over time in fine-grained scale, playing an increasingly important role in agriculture, defense, and emergency response. However, most current algorithms are still confined to describing local features and fail to incorporate a global perspective, which limits their ability to capture interactions between global features, thus usually resulting in incomplete change regions. In this paper, we proposed a Global Multi-head INteractive self-attention change Detection network (GlobalMind) to explore the implicit correlation between different surface objects and variant land cover transformations, acquiring a comprehensive understanding of the data and accurate change detection result. Firstly, a simple but effective Global Axial Segmentation (GAS) strategy is designed to expand the self-attention computation along the row space or column space of hyperspectral images, allowing the global connection with high efficiency. Secondly, with GAS, the global spatial multi-head interactive self-attention (Global-M) module is crafted to mine the abundant spatial-spectral feature involving potential correlations between the ground objects from the entire rich and complex hyperspectral space. Moreover, to acquire the accurate and complete cross-temporal changes, we devise a global temporal interactive multi-head self-attention (GlobalD) module which incorporates the relevance and variation of bi-temporal spatialspectral features, deriving the integrate potential same kind of changes in the local and global range with the combination of GAS. We perform extensive experiments on five mostly used hyperspectral datasets, and our method outperforms the state-ofthe-art algorithms with high accuracy and efficiency.

show abstract

“…Qu et al [48] introduced the graph attention network to HSI CD for the first time, which leveraged the spatial-temporal joint correlations to explore multiple features. In [49], cross-temporal attention was designed to explore the temporal change information between bi-temporal features. Ou et al [50] performed attention operations on image patches of different scales at the same time so that the central pixel to be detected in the fused feature map has a higher weight.…”

Section: Introductionmentioning

confidence: 99%

Spectral–Temporal Transformer for Hyperspectral Image Change Detection

Ding

2023

Remote Sensing

View full text Add to dashboard Cite

Deep-Learning-based (DL-based) approaches have achieved remarkable performance in hyperspectral image (HSI) change detection (CD). Convolutional Neural Networks (CNNs) are often employed to capture fine spatial features, but they do not effectively exploit the spectral sequence information. Furthermore, existing Siamese-based networks ignore the interaction of change information during feature extraction. To address this issue, we propose a novel architecture, the Spectral–Temporal Transformer (STT), which processes the HSI CD task from a completely sequential perspective. The STT concatenates feature embeddings in spectral order, establishing a global spectrum–time-receptive field that can learn different representative features between two bands regardless of spectral or temporal distance, thereby strengthening the learning of temporal change information. Via the multi-head self-attention mechanism, the STT is capable of capturing spectral–temporal features that are weighted and enriched with discriminative sequence information, such as inter-spectral correlations, variations, and time dependency. We conducted experiments on three HSI datasets, demonstrating the competitive performance of our proposed method. Specifically, the overall accuracy of the STT outperforms the second-best method by 0.08%, 0.68%, and 0.99% on the Farmland, Hermiston, and River datasets, respectively.

show abstract

CSANet: Cross-Temporal Interaction Symmetric Attention Network for Hyperspectral Image Change Detection

Cited by 41 publications

References 12 publications

Cross-receptive Focused Inference Network for Lightweight Image Super-Resolution

Cross-receptive Focused Inference Network for Lightweight Image Super-Resolution

Multi-Temporal Spatial-Spectral Comparison Network For Hyperspectral Anomalous Change Detection

Spectral–Temporal Transformer for Hyperspectral Image Change Detection

Contact Info

Product

Resources

About