Haoyuan Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Haoyuan Wang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Xuening Li
School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
3 shared publications - Yan Xu
School of Electrical and Information Engineering, Tianjin University, Tianjin, China
3 shared publications - Zhong Ji
School of Electrical and Information Engineering, Tianjin Key Laboratory of Brain-Inspired Intelligence Technology, Tianjin University, Tianjin, China
2 shared publications - Bowen Huang
School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China
1 shared publication - Chao Zhou
National Marine Environmental Monitoring Center, Dalian 116023, China
1 shared publication - Chen Hu
State Key Laboratory of Oral Diseases & National Center for Stomatology & National Clinical Research Center for Oral Diseases & Department of Oral Implantology, West China Hospital of Stomatology, Sichuan University, Chengdu, China
1 shared publication
Latest Publications
- Open Access
- 20 December 2024
- Published by MDPI AG in Electronics
- Vol. 13 (24) , 5013
Video Anomaly Detection with Hyperbolic Graph Embedding and Masked Normalizing Flows
Video anomaly detection plays a vital role in intelligent video monitoring systems. It has found extensive applications in the fields of public safety and social security. Nevertheless, the area of video anomaly detection continues to be a formidable task because of the intricate nature of actual data and the challenge of precisely identifying anomalies. Current anomaly detection approaches suffer from overpowered generalization ability, weak spatio-temporal feature extraction capability, and insufficient global information aggregation ability. Therefore, an unsupervised Hyperbolic Graph-based Normalizing Flows (HGNF) model is proposed in this paper, which is constructed with a Spatio-temporal Encoder (STEncoder) and stacked normalizing flows to reduce the overgeneralization of auto-encoder-based anomaly detection models. STEncoder consists of spatio-temporal attention and inter-frame feature aggregation. In normalizing flows, a Poincaré ball graph extractor is developed to improve the representation ability of the dynamic changes of the input data, and a masked affine coupling block is established to improve the performance of this model in global information aggregation. According to the experimental results obtained on four public datasets, HGNF achieves excellent performance and the best AUC (Area Under the Curve) score of 74.5% on the UBnormal dataset.
- 20 December 2024
- Published by MDPI AG in Electronics
- Vol. 13 (24) , 5013
Open AccessVideo Anomaly Detection with Hyperbolic Graph Embedding and Masked Normalizing Flows
Video anomaly detection plays a vital role in intelligent video monitoring systems. It has found extensive applications in the fields of public safety and social security. Nevertheless, the area of video anomaly detection continues to be a formidable task because of the intricate nature of actual data and the challenge of precisely identifying anomalies. Current anomaly detection approaches suffer from overpowered generalization ability, weak spatio-temporal feature extraction capability, and insufficient global information aggregation ability. Therefore, an unsupervised Hyperbolic Graph-based Normalizing Flows (HGNF) model is proposed in this paper, which is constructed with a Spatio-temporal Encoder (STEncoder) and stacked normalizing flows to reduce the overgeneralization of auto-encoder-based anomaly detection models. STEncoder consists of spatio-temporal attention and inter-frame feature aggregation. In normalizing flows, a Poincaré ball graph extractor is developed to improve the representation ability of the dynamic changes of the input data, and a masked affine coupling block is established to improve the performance of this model in global information aggregation. According to the experimental results obtained on four public datasets, HGNF achieves excellent performance and the best AUC (Area Under the Curve) score of 74.5% on the UBnormal dataset.
- 3 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
Depth-Vision-Decoupled Transformer With Cascaded Group Convolutional Attention for Monocular 3-D Object Detection
Monocular three-dimensional (3D) object detection is a challenging and important task in autonomous driving. In 3D object detection, depth information is crucial for determining the position, size, and posture of objects. However, in the two-dimensional (2D) images obtained by monocular cameras, depth information is compressed into a plane, which blurs the relative position and size relationships between objects. In addition, the pose changes of objects in 3D space (such as rotation, tilt, etc.) can cause changes in their projected shape, size, and position in 2D images, which increases the difficulty of detecting 3D objects from monocular 2D images. In this paper, we propose a depth-vision-decoupled transformer for monocular multi-class 3D object detection. The proposed scheme is constructed with the following novel components: (1) a Cascaded Group-Multi Scale Convolutional Attention (CG-MSCA) with multi-scale-perception and direction-perception capabilities to focus on the local characteristics of the complex scene; (2) a Decoupled-Feature-Aware TRansformer (DFTR) module, which globally decouples depth and visual features, encodes and decodes them separately to avoid the monocularly estimated inaccurate depth interfering with the model’s learning of visual information and alleviate the mismatch between stereo information and geometric shape of the object; (3) a Cross-Attention-Guided Fusion Module (CAFM) to rationalize the fusion of decoupled depth and visual features before prediction. Experiments on the KITTI, DAIR-V2X-V, and DAIR-V2X-I datasets show that our proposed method produces competitive performance compared with Transformer-based methods and other state-of-the-art (SOTA) methods. For the Car Category, our method achieves 17.04% Average Precision of three-dimensional (AP3D) for the moderate setting (the most important setting) of the KITTI dataset with an Intersection over Union (IoU) threshold of 0.7.
- 3 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-13
Depth-Vision-Decoupled Transformer With Cascaded Group Convolutional Attention for Monocular 3-D Object Detection
Monocular three-dimensional (3D) object detection is a challenging and important task in autonomous driving. In 3D object detection, depth information is crucial for determining the position, size, and posture of objects. However, in the two-dimensional (2D) images obtained by monocular cameras, depth information is compressed into a plane, which blurs the relative position and size relationships between objects. In addition, the pose changes of objects in 3D space (such as rotation, tilt, etc.) can cause changes in their projected shape, size, and position in 2D images, which increases the difficulty of detecting 3D objects from monocular 2D images. In this paper, we propose a depth-vision-decoupled transformer for monocular multi-class 3D object detection. The proposed scheme is constructed with the following novel components: (1) a Cascaded Group-Multi Scale Convolutional Attention (CG-MSCA) with multi-scale-perception and direction-perception capabilities to focus on the local characteristics of the complex scene; (2) a Decoupled-Feature-Aware TRansformer (DFTR) module, which globally decouples depth and visual features, encodes and decodes them separately to avoid the monocularly estimated inaccurate depth interfering with the model’s learning of visual information and alleviate the mismatch between stereo information and geometric shape of the object; (3) a Cross-Attention-Guided Fusion Module (CAFM) to rationalize the fusion of decoupled depth and visual features before prediction. Experiments on the KITTI, DAIR-V2X-V, and DAIR-V2X-I datasets show that our proposed method produces competitive performance compared with Transformer-based methods and other state-of-the-art (SOTA) methods. For the Car Category, our method achieves 17.04% Average Precision of three-dimensional (AP3D) for the moderate setting (the most important setting) of the KITTI dataset with an Intersection over Union (IoU) threshold of 0.7.
- 27 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-11
Global-Context-Aware Visual Odometry System With Epipolar-Geometry-Constrained Loss Function
Visual odometry (VO) plays a vital role in Simultaneous Localization and Mapping (SLAM). Most of the current learning-based VO methods utilize Convolutional Neural Network (CNN) as framework. However, CNN is weak in integrating global context information. In the design of loss function, the majority of these methods ignore the restraint relationship between translation and rotation prediction. In this work we propose an end-to-end global-context-aware visual odometry scheme with Epipolar-Geometry-constrained loss function to estimate the relative 6-DoF poses of monocular camera. The proposed scheme designs an augmented-attention-enhanced global context block on top of contextual CNN to learn the long-range dependencies and internal correlation. To overcome the problem of mutual restraint between the translation and rotation errors, an Epipolar-Geometry-constrained loss function is developed to simultaneously improve the prediction accuracy of both translation and rotation. The evaluation results on public datasets and self-collected dataset show that the proposed system outperforms state-of-the-art learning-based VO methods with a large margin.
- 27 February 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Instrumentation and Measurement
- Vol. 73 (00189456) , 1-11
Global-Context-Aware Visual Odometry System With Epipolar-Geometry-Constrained Loss Function
Visual odometry (VO) plays a vital role in Simultaneous Localization and Mapping (SLAM). Most of the current learning-based VO methods utilize Convolutional Neural Network (CNN) as framework. However, CNN is weak in integrating global context information. In the design of loss function, the majority of these methods ignore the restraint relationship between translation and rotation prediction. In this work we propose an end-to-end global-context-aware visual odometry scheme with Epipolar-Geometry-constrained loss function to estimate the relative 6-DoF poses of monocular camera. The proposed scheme designs an augmented-attention-enhanced global context block on top of contextual CNN to learn the long-range dependencies and internal correlation. To overcome the problem of mutual restraint between the translation and rotation errors, an Epipolar-Geometry-constrained loss function is developed to simultaneously improve the prediction accuracy of both translation and rotation. The evaluation results on public datasets and self-collected dataset show that the proposed system outperforms state-of-the-art learning-based VO methods with a large margin.
- 11 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 20, 5791-5794
A New Categories Identification Method based on Reliability Test in Radar Signal Recognition System
In the field of radar electronic reconnaissance, radar signal recognition is a key technology. In the actual task, part of the signals to be recognized may come from new types of emitters, which can not be identified directly by the existing recognition system. In order to get the ability to recognize new categories, it is necessary to analyze the unrecognized samples for incremental learning. In this paper, a new categories identification method based on reliability test is proposed. Firstly, an existing clustering method is used to label the unrecognized samples, and then the reliability test criteria are designed, including quantity criterion, distance criterion and frequency criterion, to screen the clustered sample points. The proposed method provides better data support for incremental learning in radar signal recognition. Simulation results show the effectiveness of the proposed method.
- 11 July 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 20, 5791-5794
A New Categories Identification Method based on Reliability Test in Radar Signal Recognition System
In the field of radar electronic reconnaissance, radar signal recognition is a key technology. In the actual task, part of the signals to be recognized may come from new types of emitters, which can not be identified directly by the existing recognition system. In order to get the ability to recognize new categories, it is necessary to analyze the unrecognized samples for incremental learning. In this paper, a new categories identification method based on reliability test is proposed. Firstly, an existing clustering method is used to label the unrecognized samples, and then the reliability test criteria are designed, including quantity criterion, distance criterion and frequency criterion, to screen the clustered sample points. The proposed method provides better data support for incremental learning in radar signal recognition. Simulation results show the effectiveness of the proposed method.
- Open Access
Multi-Granularity Modularized Network for Abstract Visual Reasoning
Abstract visual reasoning connects mental abilities to the physical world, which is a crucial factor in cognitive development. Most toddlers display sensitivity to this skill, but it is not easy for machines. Aimed at it, we focus on the Raven Progressive Matrices Test, designed to measure cognitive reasoning. Recent work designed some black-boxes to solve it in an end-to-end fashion, but they are incredibly complicated and difficult to explain. Inspired by cognitive studies, we propose a Multi-Granularity Modularized Network (MMoN) to bridge the gap between the processing of raw sensory information and symbolic reasoning. Specifically, it learns modularized reasoning functions to model the semantic rule from the visual grounding in a neuro-symbolic and semi-supervision way. To comprehensively evaluate MMoN, our experiments are conducted on the dataset of both seen and unseen reasoning rules. The result shows that MMoN is well suited for abstract visual reasoning and also explainable on the generalization test.
Open AccessMulti-Granularity Modularized Network for Abstract Visual Reasoning
Abstract visual reasoning connects mental abilities to the physical world, which is a crucial factor in cognitive development. Most toddlers display sensitivity to this skill, but it is not easy for machines. Aimed at it, we focus on the Raven Progressive Matrices Test, designed to measure cognitive reasoning. Recent work designed some black-boxes to solve it in an end-to-end fashion, but they are incredibly complicated and difficult to explain. Inspired by cognitive studies, we propose a Multi-Granularity Modularized Network (MMoN) to bridge the gap between the processing of raw sensory information and symbolic reasoning. Specifically, it learns modularized reasoning functions to model the semantic rule from the visual grounding in a neuro-symbolic and semi-supervision way. To comprehensively evaluate MMoN, our experiments are conducted on the dataset of both seen and unseen reasoning rules. The result shows that MMoN is well suited for abstract visual reasoning and also explainable on the generalization test.
- 1 May 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3944, 1-4
AFD: A feature detection method for outdoor real-time video stitching system
This paper proposes an efficient feature detection solution called AFD (Auto-adapted Feature Detection) based on FAST and LDB applied to outdoor real-time video stitching system. As FAST features are sensitive to illumination and blur, AFD comes out to optimize it. AFD detects features with local and global auto-adapted FAST threshold. Global score compensation is adopted to get good performance under different outdoor situation as well. We use a multiple grid test method to generate more distinctive descriptors. In order to reduce the computing time and increase accuracy, features are extracted in the region of interest (ROI). The feature detection with low complexity and robustness performs well in the real-time stitching system.
- 1 May 2017
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3944, 1-4
AFD: A feature detection method for outdoor real-time video stitching system
This paper proposes an efficient feature detection solution called AFD (Auto-adapted Feature Detection) based on FAST and LDB applied to outdoor real-time video stitching system. As FAST features are sensitive to illumination and blur, AFD comes out to optimize it. AFD detects features with local and global auto-adapted FAST threshold. Global score compensation is adopted to get good performance under different outdoor situation as well. We use a multiple grid test method to generate more distinctive descriptors. In order to reduce the computing time and increase accuracy, features are extracted in the region of interest (ROI). The feature detection with low complexity and robustness performs well in the real-time stitching system.