Zhiyong Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Zhiyong Liu has a total of 5 co-authors. Here are the top co-authors by number of shared publications:
- Guoqian Luo
School of Information Science and Technology, Northeast Normal University, Changchun, China
3 shared publications - Kehan Wang
School of Information Science and Technology, Northeast Normal University, Changchun, China
3 shared publications - Changming Li
School of Physics, Changchun University of Science and Technology, Changchun 130022, China
2 shared publications - Yixuan Wang
School of Information Science and Technology, Northeast Normal University, Changchun, China
2 shared publications - Yaozheng Wang
Northeast Normal University
1 shared publication
Latest Publications
- Open Access
- 2 November 2023
- Published by Springer Nature in Research Square
EB-YOLOX:A Balanced One-Stage Object Detection Model Concentrating on Global Features
With the increasing complexity of objects and the limitation of hardware resources, it is crucial to design object detection models that are both highly effective and efficient. Although YOLOX, as a leading object detector, strikes a good balance between the number of parameters and performance, it still has some design weaknesses. These include constraints in Bottleneck extracted features, insufficient feature fusion and information loss in FPN (Feature Pyramid Networks), and an uneven trade-off between performance and efficiency in the detection head. These limitations affect the equilibrium between the number of parameters and model performance. To address these issues, this paper presents an object detection model known as EB-YOLOX (Emphasized Balance YOLOX), which incorporates three key elements: Multidirectional Feature Extraction Bottleneck, Decoupled bidirectional Feature Pyramid Network, and Efficient Decoupling head network. With these components, EB-YOLOX effectively tackles model performance challenges and undergoes a comprehensive evaluation on the MS COCO dataset. Experimental results indicate that EB-YOLOX outperforms YOLOX-S by 1.8% in Average Precision (AP) and achieves significant performance improvements across various scenarios, highlighting its excellent balance between the number of parameters and performance.
- 2 November 2023
- Published by Springer Nature in Research Square
Open AccessEB-YOLOX:A Balanced One-Stage Object Detection Model Concentrating on Global Features
With the increasing complexity of objects and the limitation of hardware resources, it is crucial to design object detection models that are both highly effective and efficient. Although YOLOX, as a leading object detector, strikes a good balance between the number of parameters and performance, it still has some design weaknesses. These include constraints in Bottleneck extracted features, insufficient feature fusion and information loss in FPN (Feature Pyramid Networks), and an uneven trade-off between performance and efficiency in the detection head. These limitations affect the equilibrium between the number of parameters and model performance. To address these issues, this paper presents an object detection model known as EB-YOLOX (Emphasized Balance YOLOX), which incorporates three key elements: Multidirectional Feature Extraction Bottleneck, Decoupled bidirectional Feature Pyramid Network, and Efficient Decoupling head network. With these components, EB-YOLOX effectively tackles model performance challenges and undergoes a comprehensive evaluation on the MS COCO dataset. Experimental results indicate that EB-YOLOX outperforms YOLOX-S by 1.8% in Average Precision (AP) and achieves significant performance improvements across various scenarios, highlighting its excellent balance between the number of parameters and performance.
- Open Access
- 10 July 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (14) , 8037
TIG-DETR: Enhancing Texture Preservation and Information Interaction for Target Detection
FPN (Feature Pyramid Network) and transformer-based target detectors are commonly employed in target detection tasks. However, these approaches suffer from design flaws that restrict their performance. To overcome these limitations, we proposed TIG-DETR (Texturized Instance Guidance DETR), a novel target detection model. TIG-DETR comprises a backbone network, TE-FPN (Texture-Enhanced FPN), and an enhanced DETR detector. TE-FPN addresses the issue of texture information loss in FPN by utilizing a bottom-up architecture, Lightweight Feature-wise Attention, and Feature-wise Attention. These components effectively compensate for texture information loss, mitigate the confounding effect of cross-scale fusion, and enhance the final output features. Additionally, we introduced the Instance Based Advanced Guidance Module in the DETR-based detector to tackle the weak detection of larger objects caused by the limitations of window interactions in Shifted Window-based Self-Attention. By incorporating TE-FPN instead of FPN in Faster RCNN and employing ResNet-50 as the backbone network, we observed an improvement of 1.9 AP in average accuracy. By introducing the Instance-Based Advanced Guidance Module, the average accuracy of the DETR-based target detector has been improved by 0.4 AP. TIG-DETR achieves an impressive average accuracy of 44.1% with ResNet-50 as the backbone network.
- 10 July 2023
- Published by MDPI AG in Applied Sciences
- Vol. 13 (14) , 8037
Open AccessTIG-DETR: Enhancing Texture Preservation and Information Interaction for Target Detection
FPN (Feature Pyramid Network) and transformer-based target detectors are commonly employed in target detection tasks. However, these approaches suffer from design flaws that restrict their performance. To overcome these limitations, we proposed TIG-DETR (Texturized Instance Guidance DETR), a novel target detection model. TIG-DETR comprises a backbone network, TE-FPN (Texture-Enhanced FPN), and an enhanced DETR detector. TE-FPN addresses the issue of texture information loss in FPN by utilizing a bottom-up architecture, Lightweight Feature-wise Attention, and Feature-wise Attention. These components effectively compensate for texture information loss, mitigate the confounding effect of cross-scale fusion, and enhance the final output features. Additionally, we introduced the Instance Based Advanced Guidance Module in the DETR-based detector to tackle the weak detection of larger objects caused by the limitations of window interactions in Shifted Window-based Self-Attention. By incorporating TE-FPN instead of FPN in Faster RCNN and employing ResNet-50 as the backbone network, we observed an improvement of 1.9 AP in average accuracy. By introducing the Instance-Based Advanced Guidance Module, the average accuracy of the DETR-based target detector has been improved by 0.4 AP. TIG-DETR achieves an impressive average accuracy of 44.1% with ResNet-50 as the backbone network.
- Open Access
- 15 June 2023
- Published by MDPI AG in Preprints.org
TIG-DETR: Enhancing Texture Preservation and Information Interaction for Target Detection
In practical applications, the detection of objects with various sizes is a common requirement for most detectors. The feature pyramid network (FPN) is widely adopted as a framework to address this challenge. The field is witnessing an increasing number of transformer-based target detectors due to the widespread adoption of transformer technology. This paper initially examines the design flaws in FPN and transformer-based target detectors, followed by the introduction of a new transformer-based approach called Texturized Instance Guidance (TIG-DETR) to address these issues. Specifically, TIG-DETR comprises a backbone network, a new pyramidal structure known as Texture-Enhanced FPN (TE-FPN), and an enhanced DETR detector.The TE-FPN is composed of three components: a bottom-up pathway for enhancing texture information in the feature map, a lightweight attention module to address confounding effects resulting from cross-scale fusion, and a standard attention module to enhance the final output features.The improved DETR detector utilizes Shifted Window based Self-Attention to replace the multi-headed self-attention module in DETR, thereby accelerating model convergence. Moreover, it incorporates an Instance Based Advanced Guidance Module to enhance instance perception in the image by employing a pre-local self-attentive mechanism for recognizing larger instances. By employing TE-FPN instead of FPN in Faster RCNN with Resnet-50 as the backbone network, we achieve a 1.9% improvement in average accuracy. TIG-DETR achieves an average accuracy of 44.1 with Resnet-50 as the backbone network.
- 15 June 2023
- Published by MDPI AG in Preprints.org
Open AccessTIG-DETR: Enhancing Texture Preservation and Information Interaction for Target Detection
In practical applications, the detection of objects with various sizes is a common requirement for most detectors. The feature pyramid network (FPN) is widely adopted as a framework to address this challenge. The field is witnessing an increasing number of transformer-based target detectors due to the widespread adoption of transformer technology. This paper initially examines the design flaws in FPN and transformer-based target detectors, followed by the introduction of a new transformer-based approach called Texturized Instance Guidance (TIG-DETR) to address these issues. Specifically, TIG-DETR comprises a backbone network, a new pyramidal structure known as Texture-Enhanced FPN (TE-FPN), and an enhanced DETR detector.The TE-FPN is composed of three components: a bottom-up pathway for enhancing texture information in the feature map, a lightweight attention module to address confounding effects resulting from cross-scale fusion, and a standard attention module to enhance the final output features.The improved DETR detector utilizes Shifted Window based Self-Attention to replace the multi-headed self-attention module in DETR, thereby accelerating model convergence. Moreover, it incorporates an Instance Based Advanced Guidance Module to enhance instance perception in the image by employing a pre-local self-attentive mechanism for recognizing larger instances. By employing TE-FPN instead of FPN in Faster RCNN with Resnet-50 as the backbone network, we achieve a 1.9% improvement in average accuracy. TIG-DETR achieves an average accuracy of 44.1 with Resnet-50 as the backbone network.