Zengbin Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zengbin Wang has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Man Zhang
Beijing University of Posts and Telecommunications
5 shared publications - Xu Liu
Watrix Technology Company Ltd., Beijing, China
4 shared publications - Yongzhen Huang
Beijing Normal University, Beijing, China
4 shared publications - Chunshui Cao
Watrix Technology Company Ltd., Beijing, China
3 shared publications - Saihui Hou
Beijing Normal University, Beijing, China
3 shared publications - Junjie Li
Beijing University of Posts and Telecommunications, Haidian, Beijing, China
2 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4817-4825
Restoring Semantic Fidelity for Training-Free High-Resolution Image Generation in Diffusion Transformers
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4817-4825
Open AccessRestoring Semantic Fidelity for Training-Free High-Resolution Image Generation in Diffusion Transformers
- Open Access
Learning Geometric Invariance for Gait Recognition
The goal of gait recognition is to extract identity-invariant features of an individual under various gait conditions, e.g., cross-view and cross-clothing. Most gait models strive to implicitly learn the common traits across different gait conditions in a data-driven manner to pull different gait conditions closer for recognition. However, relatively few studies have explicitly explored the inherent relations between different gait conditions. For this purpose, we attempt to establish connections among different gait conditions and propose a new perspective to achieve gait recognition: variations in different gait conditions can be approximately viewed as a combination of geometric transformations. In this case, all we need is to determine the types of geometric transformations and achieve geometric invariance, then identity invariance naturally follows. As an initial attempt, we explore three common geometric transformations (i.e., Reflect, Rotate, and Scale) and design a $\mathcal{R}$eflect-$\mathcal{R}$otate-$\mathcal{S}$cale invariance learning framework, named ${\mathcal{RRS}}$-Gait. Specifically, it first flexibly adjusts the convolution kernel based on the specific geometric transformations to achieve approximate feature equivariance. Then these three equivariant-aware features are respectively fed into a global pooling operation for final invariance-aware learning. Extensive experiments on four popular gait datasets (Gait3D, GREW, CCPG, SUSTech1K) show superior performance across various gait conditions.
Open AccessLearning Geometric Invariance for Gait Recognition
The goal of gait recognition is to extract identity-invariant features of an individual under various gait conditions, e.g., cross-view and cross-clothing. Most gait models strive to implicitly learn the common traits across different gait conditions in a data-driven manner to pull different gait conditions closer for recognition. However, relatively few studies have explicitly explored the inherent relations between different gait conditions. For this purpose, we attempt to establish connections among different gait conditions and propose a new perspective to achieve gait recognition: variations in different gait conditions can be approximately viewed as a combination of geometric transformations. In this case, all we need is to determine the types of geometric transformations and achieve geometric invariance, then identity invariance naturally follows. As an initial attempt, we explore three common geometric transformations (i.e., Reflect, Rotate, and Scale) and design a $\mathcal{R}$eflect-$\mathcal{R}$otate-$\mathcal{S}$cale invariance learning framework, named ${\mathcal{RRS}}$-Gait. Specifically, it first flexibly adjusts the convolution kernel based on the specific geometric transformations to achieve approximate feature equivariance. Then these three equivariant-aware features are respectively fed into a global pooling operation for final invariance-aware learning. Extensive experiments on four popular gait datasets (Gait3D, GREW, CCPG, SUSTech1K) show superior performance across various gait conditions.
- 10 October 2024
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 133 (4) , 1549-1563
Edge-Oriented Adversarial Attack for Deep Gait Recognition
- 10 October 2024
- Published by Springer Nature in International Journal of Computer Vision
- Vol. 133 (4) , 1549-1563
Edge-Oriented Adversarial Attack for Deep Gait Recognition
- Open Access
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2305-2314
LandmarkGait: Intrinsic Human Parsing for Gait Recognition
Gait recognition is an emerging biometric technology for identifying pedestrians based on their unique walking patterns. In past gait recognition, global-based methods are inadequate to meet the growing demand for accuracy, while commonly used part-based methods provided coarse and inaccurate feature representation for specific body parts. Human parsing appears to be a better option for accurately representing specific and complete body parts in gait recognition. However, its practical application in gait recognition is often hindered by missing RGB modality, lack of annotated body parts, and difficulty in balancing parsing quantity and quality. To address this issue, we propose LandmarkGait, an accessible and alternative parsing-based solution for gait recognition. LandmarkGait introduces an unsupervised landmark discovery network to transform the dense silhouette into a finite set of landmarks with remarkable consistency across various conditions. By grouping landmarks subsets corresponding to distinct body part regions, following a reconstruction task and further refinement from high-quality input silhouettes, we can directly obtain fine-grained parsing results from original binary silhouettes in an unsupervised manner. Moreover, we also develop a multi-scale feature extractor that simultaneously captures global and parsing feature representations based on the integrity and flexibility of specific body parts. Extensive experiments demonstrate that our LandmarkGait can extract more stable features and exhibit significant performance improvement under all conditions, especially in various dressing conditions. Code is available at https://github.com/wzb-bupt/LandmarkGait.
- 26 October 2023
- Published by Association for Computing Machinery (ACM)
- p. 2305-2314
Open AccessLandmarkGait: Intrinsic Human Parsing for Gait Recognition
Gait recognition is an emerging biometric technology for identifying pedestrians based on their unique walking patterns. In past gait recognition, global-based methods are inadequate to meet the growing demand for accuracy, while commonly used part-based methods provided coarse and inaccurate feature representation for specific body parts. Human parsing appears to be a better option for accurately representing specific and complete body parts in gait recognition. However, its practical application in gait recognition is often hindered by missing RGB modality, lack of annotated body parts, and difficulty in balancing parsing quantity and quality. To address this issue, we propose LandmarkGait, an accessible and alternative parsing-based solution for gait recognition. LandmarkGait introduces an unsupervised landmark discovery network to transform the dense silhouette into a finite set of landmarks with remarkable consistency across various conditions. By grouping landmarks subsets corresponding to distinct body part regions, following a reconstruction task and further refinement from high-quality input silhouettes, we can directly obtain fine-grained parsing results from original binary silhouettes in an unsupervised manner. Moreover, we also develop a multi-scale feature extractor that simultaneously captures global and parsing feature representations based on the integrity and flexibility of specific body parts. Extensive experiments demonstrate that our LandmarkGait can extract more stable features and exhibit significant performance improvement under all conditions, especially in various dressing conditions. Code is available at https://github.com/wzb-bupt/LandmarkGait.
- 19 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 4736-4748
GaitParsing: Human Semantic Parsing for Gait Recognition
Gait recognition is a soft biotechnology to identify pedestrians observed from different camera views based on specific walking patterns. However, various dressing and wearing conditions bring great challenges to realistic gait recognition. Most existing methods take holistic gait silhouette as input and focus on local areas through horizontal strip division or attention map. We consider that this processing may contain mixed or incomplete information about multiple body parts so that gait information is misused or underutilized. In this paper, we propose a parsing-guided framework for gait recognition, named GaitParsing , which explores human semantic parsing to dissect human body into a set of specific and complete body parts. Correspondingly, a simple yet effective dual-branch feature extraction network is adopted to process holistic gait and distinct body parts. To maximize the use of highly discriminated gait frames, we propose a self-occlusion frame assessment to measure the self-occlusion in a gait sequence. Since there is no human parsing modality in current gait datasets, we further develop a general human parsing pipeline specifically tailored for gait datasets. This single training enables widespread application across various gait datasets. Extensive experiments with ablation analyses demonstrate competitive performance even in the most challenging conditions, e.g., Cloth-Changing (CC+5.9%). Especially, It is gratifying to see that our model can be easily applied to existing methods and significantly outperform the original architecture, even without much modification. Code is available at https://github.com/wzb-bupt/GaitParsing
- 19 October 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 26 (15209210) , 4736-4748
GaitParsing: Human Semantic Parsing for Gait Recognition
Gait recognition is a soft biotechnology to identify pedestrians observed from different camera views based on specific walking patterns. However, various dressing and wearing conditions bring great challenges to realistic gait recognition. Most existing methods take holistic gait silhouette as input and focus on local areas through horizontal strip division or attention map. We consider that this processing may contain mixed or incomplete information about multiple body parts so that gait information is misused or underutilized. In this paper, we propose a parsing-guided framework for gait recognition, named GaitParsing , which explores human semantic parsing to dissect human body into a set of specific and complete body parts. Correspondingly, a simple yet effective dual-branch feature extraction network is adopted to process holistic gait and distinct body parts. To maximize the use of highly discriminated gait frames, we propose a self-occlusion frame assessment to measure the self-occlusion in a gait sequence. Since there is no human parsing modality in current gait datasets, we further develop a general human parsing pipeline specifically tailored for gait datasets. This single training enables widespread application across various gait datasets. Extensive experiments with ablation analyses demonstrate competitive performance even in the most challenging conditions, e.g., Cloth-Changing (CC+5.9%). Especially, It is gratifying to see that our model can be easily applied to existing methods and significantly outperform the original architecture, even without much modification. Code is available at https://github.com/wzb-bupt/GaitParsing
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 527-530
Simulation Study on Wind Load and Mechanical Characteristics of High Voltage Dis-Connector
The high-voltage dis-connector, as a key substation equipment in the power grid structure, is vitally important to the quality of the power grid operation. Due to the geographic location of Guangdong Province of China, many strong typhoons have caused great damage to power facilities in recent years. And the loss of the dis-connector is the most serious, which severely affects the reliability of the power supply. However, the research on the wind resistance of the dis-connector is still few, so the improvement work of the device is quite difficult. The phenomenon of strong typhoon destroying the dis-connector mainly includes the fracture of post insulator and the conductive arm. In the study of the mechanical properties of post insulator and the conductive arm, almost all wind loads are treated as a concentrated force, but the surfaces of the post insulator and the conductive arm are actually subjected to a continuous wind pressure distribution. The magnitude of the wind load and the error of its application method will inevitably lead to inaccurate stress and strain data obtained in the calculation of the strength of the post insulator and the conductive arm, so that the post insulator and the conductive arm cannot be accurately designed. In this paper, the wind resistance of the dis-connector are studied by using the fluid-structural coupling analysis method. Firstly, the hydrodynamic RNG k-ε turbulence model is used to simulate the wind field around the dis-connector, so as to obtain the accurate result of the wind pressure distribution on the surface of the dis-connector. Then the surface pressure of the dis-connector calculated by hydrodynamics simulation is introduced into the structural module of Ansys software as the applied load. The finite element method is used to simulate and calculate the stress and strain of the dis-connector, so as to obtain the mechanical characteristics of the dis-connector under strong wind. In addition, this paper analyzes the effects of different wind speeds and different wind directions on the wind load applied to the surface of dis-connector, and provide a basis for wind resistance design of dis-connector.
- 1 October 2019
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 527-530
Simulation Study on Wind Load and Mechanical Characteristics of High Voltage Dis-Connector
The high-voltage dis-connector, as a key substation equipment in the power grid structure, is vitally important to the quality of the power grid operation. Due to the geographic location of Guangdong Province of China, many strong typhoons have caused great damage to power facilities in recent years. And the loss of the dis-connector is the most serious, which severely affects the reliability of the power supply. However, the research on the wind resistance of the dis-connector is still few, so the improvement work of the device is quite difficult. The phenomenon of strong typhoon destroying the dis-connector mainly includes the fracture of post insulator and the conductive arm. In the study of the mechanical properties of post insulator and the conductive arm, almost all wind loads are treated as a concentrated force, but the surfaces of the post insulator and the conductive arm are actually subjected to a continuous wind pressure distribution. The magnitude of the wind load and the error of its application method will inevitably lead to inaccurate stress and strain data obtained in the calculation of the strength of the post insulator and the conductive arm, so that the post insulator and the conductive arm cannot be accurately designed. In this paper, the wind resistance of the dis-connector are studied by using the fluid-structural coupling analysis method. Firstly, the hydrodynamic RNG k-ε turbulence model is used to simulate the wind field around the dis-connector, so as to obtain the accurate result of the wind pressure distribution on the surface of the dis-connector. Then the surface pressure of the dis-connector calculated by hydrodynamics simulation is introduced into the structural module of Ansys software as the applied load. The finite element method is used to simulate and calculate the stress and strain of the dis-connector, so as to obtain the mechanical characteristics of the dis-connector under strong wind. In addition, this paper analyzes the effects of different wind speeds and different wind directions on the wind load applied to the surface of dis-connector, and provide a basis for wind resistance design of dis-connector.