NSTL国家科技图书文献中心

1. Hierarchical integration transformer for marine snow removal NSTL国家科技图书文献中心

Haoru Zhao | Binghua Yin ... - 《Fourth International Conference on Computer Vision,Application,and Algorithm (CVAA 2024)》 - International Conference on Computer Vision,Application,and Algorithm - 2025, - 134861L.1～134861L.5 - 共5页

摘要：Transformer has demonstrated excellent | , Transformer* tend to overlook local image details, and their* | hierarchical integrated Transformer* for marine snow removal* | . Specifically, we leverage the Transformer* to capture global* | information across multiple levels from Transformer* and CNN*

关键词： Transformer | CNN | Global-Local integration

2. Spiking Wavelet Transformer NSTL国家科技图书文献中心

Yuetong Fang | Ziqing Wang ... - 《Computer Vision - ECCV 2024,Part LXXVI》 - European Conference on Computer Vision - 2025, - 19～37 - 共19页

摘要： Spiking Wavelet Transformer* (SWformer), an attention* | /Spiking-Wavelet-Transformer. | . Incorporating Transformers* with SNNs has shown promise for* | Transformers in capturing high-frequency visual components | dataset compared to vanilla Spiking Transformers. The

关键词： Spiking neural networks | Wavelet transform | Vision transformer | Event-based vision

3. Selective Information Flow for Transformer Tracking NSTL国家科技图书文献中心

Janani Kugarajeevan | Kokul Thanikasalam ... - 《Expert Systems with Application》 - 2025,259(Jan.) - 125381.1～125381.14 - 共14页

摘要：Fully transformer-based one-stream trackers | of the template. Moreover, previous transformer | identified limitations in transformer* tracking, we propose* | have demonstrated outstanding performance on | challenging benchmark datasets over the past three years

关键词： Transformer Tracking | Visual Object Tracking | Vision Transformer | One-stream Tracking

4. APT: Alarm Prediction Transformer NSTL国家科技图书文献中心

Nika Strem | Devendra Singh Dhami ... - 《Expert Systems with Application》 - 2025,261(Feb.) - 125521.1～125521.13 - 共13页 - 被引量：1

摘要： introduce Alarm Prediction Transformer* (APT), a multimodal* | Transformer-based machine learning model for early alarm | Distributed control systems (DCS) are | essential to operate complex industrial processes. A major | part of a DCS is the alarm system, which helps plant

关键词： Machine learning | Deep learning | Industrial processes | Alarm management | Multimodal transformer | Multimodal fusion

5. Dynamic Window Transformer for Image Super-Resolution NSTL国家科技图书文献中心

Zheng Xie | Zhongxun Wang ... - 《Computer Vision - ACCV 2024,Part VII》 - Asian Conference on Computer Vision - 2025, - 435～449 - 共15页

摘要： (LR) inputs. Recently, Swin-Transformer-based models | integration. However, existing Swin-Transformer-based SR | Dynamic Window Transformer* (DWT), a simple but novel* | Image super-resolution (SR) reconstruction is | a critical task in image processing that aims to

关键词： Super-Resolution | Dynamic-Window self-Attention | Transformer

6. Transformer for Object Re-identification: A Survey NSTL国家科技图书文献中心

Ye, Mang | Chen, Shuoyi ... - 《International Journal of Computer Vision》 - 2025,133(5) - 2410～2440 - 共31页

摘要： of studies delving deeper into Transformer-based Re | review and in-depth analysis of the Transformer-based | Transformer in addressing a multitude of challenges across | Re-ID, we propose a new Transformer* baseline* | applicability of Transformer* for this task and facilitate*

关键词： Object Re-Identification | Transformer | Survey | Person Re-Identification | Deep Learning

7. Texture graph transformer for prostate cancer classification NSTL国家科技图书文献中心

Zhang G. | Gao L.... - 《Biomedical signal processing and control》 - 2025,99(Jan.) - 1.1～1.8 - 共8页

摘要： called the texture graph transformer, which combines | transformer adeptly captures intricate texture details and | © 2024Prostate cancer classification plays a | pivotal role in the diagnosis and treatment of this | disease. In this paper, we present a novel approach

关键词： Graph convolution | Prostate cancer | Texture feature | Transformer

8. Analysis-by-Synthesis Transformer for Single-View 3D Reconstruction NSTL国家科技图书文献中心

Dian Jia | Xiaoqian Ruan ... - 《Computer Vision - ECCV 2024,Part XXI》 - European Conference on Computer Vision - 2025, - 259～277 - 共19页

摘要：-Synthesis Transformer* that addresses these limitations in* | consists of a Shape Transformer and a Texture Transformer | . The Shape Transformer* employs learnable shape* | recovering occluded vertices. The Texture Transformer | Deep learning approaches have made significant

关键词： Single-view 3D reconstruction | Shape transformer | Texture transformer

9. Mask-Aware Transformer for Crowd Counting NSTL国家科技图书文献中心

Sarah Jad | Marwan Torki ... - 《Pattern Recognition,Part XVIII》 - International Conference on Pattern Recognition - 2025, - 16～30 - 共15页

摘要： paper is a mask-aware transformer-based network for | transformer. Our proposed Mask-aware transformer (M-Trans | Crowd counting problem is a challenging task | in computer vision and image analysis. It has many | applications in the real world such as crowd management

关键词： Crowd counting | Vision transformer | Mask-aware model | Density map

10. Transformer learning-based efficient MIMO detection method NSTL国家科技图书文献中心

Burera | Saleem Ahmed ... - 《Physical communication》 - 2025,70(Jun.) - 102637.1～102637.6 - 共6页

摘要： block of a transformer* learning approach that has been* | Signal detection for multiple-input-multiple | -output (MIMO) systems is a challenging problem due to | its computational complexity. The conventional | algorithms used in this problem often are either

关键词： Transformer | CNN | Global-Local integration

摘要： Spiking Wavelet Transformer* (SWformer), an attention* | /Spiking-Wavelet-Transformer. | . Incorporating Transformers* with SNNs has shown promise for* | Transformers in capturing high-frequency visual components | dataset compared to vanilla Spiking Transformers. The

关键词： Spiking neural networks | Wavelet transform | Vision transformer | Event-based vision

摘要：Fully transformer-based one-stream trackers | of the template. Moreover, previous transformer | identified limitations in transformer* tracking, we propose* | have demonstrated outstanding performance on | challenging benchmark datasets over the past three years

关键词： Transformer Tracking | Visual Object Tracking | Vision Transformer | One-stream Tracking

摘要： introduce Alarm Prediction Transformer* (APT), a multimodal* | Transformer-based machine learning model for early alarm | Distributed control systems (DCS) are | essential to operate complex industrial processes. A major | part of a DCS is the alarm system, which helps plant

关键词： Machine learning | Deep learning | Industrial processes | Alarm management | Multimodal transformer | Multimodal fusion

摘要： (LR) inputs. Recently, Swin-Transformer-based models | integration. However, existing Swin-Transformer-based SR | Dynamic Window Transformer* (DWT), a simple but novel* | Image super-resolution (SR) reconstruction is | a critical task in image processing that aims to

关键词： Super-Resolution | Dynamic-Window self-Attention | Transformer

摘要： of studies delving deeper into Transformer-based Re | review and in-depth analysis of the Transformer-based | Transformer in addressing a multitude of challenges across | Re-ID, we propose a new Transformer* baseline* | applicability of Transformer* for this task and facilitate*

关键词： Object Re-Identification | Transformer | Survey | Person Re-Identification | Deep Learning

摘要： called the texture graph transformer, which combines | transformer adeptly captures intricate texture details and | © 2024Prostate cancer classification plays a | pivotal role in the diagnosis and treatment of this | disease. In this paper, we present a novel approach

关键词： Graph convolution | Prostate cancer | Texture feature | Transformer

摘要：-Synthesis Transformer* that addresses these limitations in* | consists of a Shape Transformer and a Texture Transformer | . The Shape Transformer* employs learnable shape* | recovering occluded vertices. The Texture Transformer | Deep learning approaches have made significant

关键词： Single-view 3D reconstruction | Shape transformer | Texture transformer

摘要： paper is a mask-aware transformer-based network for | transformer. Our proposed Mask-aware transformer (M-Trans | Crowd counting problem is a challenging task | in computer vision and image analysis. It has many | applications in the real world such as crowd management

关键词： Crowd counting | Vision transformer | Mask-aware model | Density map

摘要： block of a transformer* learning approach that has been* | Signal detection for multiple-input-multiple | -output (MIMO) systems is a challenging problem due to | its computational complexity. The conventional | algorithms used in this problem often are either

关键词： MIMO | Signal detection | Transformer | Deep neural network

4008-161-200 800-990-8900

国家科技图书文献中心

摘要：Transformer has demonstrated excellent | , Transformer tend to overlook local image details, and their | hierarchical integrated Transformer for marine snow removal | . Specifically, we leverage the Transformer to capture global | information across multiple levels from Transformer and CNN

关键词： Transformer | CNN | Global-Local integration

摘要： Spiking Wavelet Transformer (SWformer), an attention | /Spiking-Wavelet-Transformer. | . Incorporating Transformers with SNNs has shown promise for | Transformers in capturing high-frequency visual components | dataset compared to vanilla Spiking Transformers. The

关键词： Spiking neural networks | Wavelet transform | Vision transformer | Event-based vision

摘要：Fully transformer-based one-stream trackers | of the template. Moreover, previous transformer | identified limitations in transformer tracking, we propose | have demonstrated outstanding performance on | challenging benchmark datasets over the past three years

关键词： Transformer Tracking | Visual Object Tracking | Vision Transformer | One-stream Tracking

摘要： introduce Alarm Prediction Transformer (APT), a multimodal | Transformer-based machine learning model for early alarm | Distributed control systems (DCS) are | essential to operate complex industrial processes. A major | part of a DCS is the alarm system, which helps plant

关键词： Machine learning | Deep learning | Industrial processes | Alarm management | Multimodal transformer | Multimodal fusion

摘要： (LR) inputs. Recently, Swin-Transformer-based models | integration. However, existing Swin-Transformer-based SR | Dynamic Window Transformer (DWT), a simple but novel | Image super-resolution (SR) reconstruction is | a critical task in image processing that aims to

关键词： Super-Resolution | Dynamic-Window self-Attention | Transformer

摘要： of studies delving deeper into Transformer-based Re | review and in-depth analysis of the Transformer-based | Transformer in addressing a multitude of challenges across | Re-ID, we propose a new Transformer baseline | applicability of Transformer for this task and facilitate

关键词： Object Re-Identification | Transformer | Survey | Person Re-Identification | Deep Learning

摘要： called the texture graph transformer, which combines | transformer adeptly captures intricate texture details and | © 2024Prostate cancer classification plays a | pivotal role in the diagnosis and treatment of this | disease. In this paper, we present a novel approach

关键词： Graph convolution | Prostate cancer | Texture feature | Transformer

摘要：-Synthesis Transformer that addresses these limitations in | consists of a Shape Transformer and a Texture Transformer | . The Shape Transformer employs learnable shape | recovering occluded vertices. The Texture Transformer | Deep learning approaches have made significant

关键词： Single-view 3D reconstruction | Shape transformer | Texture transformer

摘要： paper is a mask-aware transformer-based network for | transformer. Our proposed Mask-aware transformer (M-Trans | Crowd counting problem is a challenging task | in computer vision and image analysis. It has many | applications in the real world such as crowd management

关键词： Crowd counting | Vision transformer | Mask-aware model | Density map

摘要： block of a transformer learning approach that has been | Signal detection for multiple-input-multiple | -output (MIMO) systems is a challenging problem due to | its computational complexity. The conventional | algorithms used in this problem often are either

关键词： MIMO | Signal detection | Transformer | Deep neural network

4008-161-200 800-990-8900

国家科技图书文献中心

摘要： Spiking Wavelet Transformer* (SWformer), an attention* | /Spiking-Wavelet-Transformer. | . Incorporating Transformers* with SNNs has shown promise for* | Transformers in capturing high-frequency visual components | dataset compared to vanilla Spiking Transformers. The

摘要：Fully transformer-based one-stream trackers | of the template. Moreover, previous transformer | identified limitations in transformer* tracking, we propose* | have demonstrated outstanding performance on | challenging benchmark datasets over the past three years

摘要： introduce Alarm Prediction Transformer* (APT), a multimodal* | Transformer-based machine learning model for early alarm | Distributed control systems (DCS) are | essential to operate complex industrial processes. A major | part of a DCS is the alarm system, which helps plant

摘要： (LR) inputs. Recently, Swin-Transformer-based models | integration. However, existing Swin-Transformer-based SR | Dynamic Window Transformer* (DWT), a simple but novel* | Image super-resolution (SR) reconstruction is | a critical task in image processing that aims to

摘要： of studies delving deeper into Transformer-based Re | review and in-depth analysis of the Transformer-based | Transformer in addressing a multitude of challenges across | Re-ID, we propose a new Transformer* baseline* | applicability of Transformer* for this task and facilitate*

摘要：-Synthesis Transformer* that addresses these limitations in* | consists of a Shape Transformer and a Texture Transformer | . The Shape Transformer* employs learnable shape* | recovering occluded vertices. The Texture Transformer | Deep learning approaches have made significant

摘要： block of a transformer* learning approach that has been* | Signal detection for multiple-input-multiple | -output (MIMO) systems is a challenging problem due to | its computational complexity. The conventional | algorithms used in this problem often are either