Zhaozhong Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhaozhong Wang has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Binglu Wang
College of Information and Control Engineering, Xian University of Architecture and Technology, Xian, China
5 shared publications - Bin Fan
School of Electronics and Information, Northwestern Polytechnical University, Xian, China
1 shared publication - Chenxi Guo
School of Astronautics, Northwestern Polytechnical University, Xian, China
1 shared publication Dazhao Zhu
YZQ Instruments Co., Ltd., Hangzhou, China
1 shared publicationDian Shao
1 shared publication- Guangyu Guo
DAMO Academy, Alibaba group, Hangzhou, Zhejiang, China and College of Computer Science and Technology, Zhejiang University, Hangzhou, Zhejiang, China
1 shared publication
Latest Publications
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (11) , 16100-16113
SCVM3D: A Hybrid Backbone With Sparse Convolutions and Mamba-2 for Voxel-Based 3D Object Detection
- 1 November 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Circuits and Systems for Video Technology
- Vol. 36 (11) , 16100-16113
SCVM3D: A Hybrid Backbone With Sparse Convolutions and Mamba-2 for Voxel-Based 3D Object Detection
- 15 April 2026
- Published by Springer Nature in Machine Intelligence Research
- Vol. 23 (5) , 1052-1067
SAMDistill: SAM-based Spatial-temporal Distillation for Robust 3D Object Detection
- 15 April 2026
- Published by Springer Nature in Machine Intelligence Research
- Vol. 23 (5) , 1052-1067
SAMDistill: SAM-based Spatial-temporal Distillation for Robust 3D Object Detection
- Open Access
- 2 February 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Acoustic shape-morphing micromachines
Shape-morphing mechanisms help organisms adapt to their environment and have inspired applications in advanced systems. Designing microrobots that are both miniaturized and capable of fast shape changes is challenging, as performance can be weakened at small scales. Ultrasound offers advantages such as fast response, repeatability, and programmability, making it suitable for enabling shape-morphing microrobots. Here, we introduce an acoustic micromachine composed of two microbubbles connected by a microhinge. Acoustic-field excitation generates interaction forces between the bubbles, enabling complete micromachine deformation within milliseconds. We also present design principles for programmable acoustic deformation, enabling both forward and inverse design, precise control, and information storage. By tuning the excitation amplitude, the micromachine can switch between multiple modes. As proof of concept, microlotus and microbird structures are demonstrated with controllable and stable performance.
- 2 February 2026
- Published by Springer Nature in Nature Communications
- Vol. 17 (1)
Open AccessAcoustic shape-morphing micromachines
Shape-morphing mechanisms help organisms adapt to their environment and have inspired applications in advanced systems. Designing microrobots that are both miniaturized and capable of fast shape changes is challenging, as performance can be weakened at small scales. Ultrasound offers advantages such as fast response, repeatability, and programmability, making it suitable for enabling shape-morphing microrobots. Here, we introduce an acoustic micromachine composed of two microbubbles connected by a microhinge. Acoustic-field excitation generates interaction forces between the bubbles, enabling complete micromachine deformation within milliseconds. We also present design principles for programmable acoustic deformation, enabling both forward and inverse design, precise control, and information storage. By tuning the excitation amplitude, the micromachine can switch between multiple modes. As proof of concept, microlotus and microbird structures are demonstrated with controllable and stable performance.
- 3 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 1689-1703
TransGOP-R: Transformer-Based Real-World Gaze Object Prediction
The goal of gaze object prediction (GOP) is to predict human gaze objects and categories. However, existing methods require additional head priors or filter the results before evaluation, which is an obstacle for real-world applications. To this end, this paper proposes a Transformer-based Gaze Object Prediction under Real-world setting (TransGOP-R), which does not rely on any head prior input and evaluates end-to-end. We first design a head location module to generate human head location information from a head query. Then, an error analysis demonstrates that the primary error source of the existing GOP model is in gaze estimation, which is caused by the difficulty in predicting gaze points by directly regressing heatmaps. Therefore, we introduce cone prediction into the model training stage, allowing the middle-layer features of the gaze regressor to build the relationship between the target human and objects before regressing the gaze point. An oriented gradient mechanism is proposed in this process to ensure the object detection performance is not affected by cone information. Finally, we conducted very detailed and sufficient experiments to verify the superiority of our method on the GOO-Synth and GOO-Real datasets. At the same time, we also achieve advantages compared to the human-target gaze estimation methods on the GazeFollowing, VideoAttentionTarget, and ChildPlay datasets.
- 3 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Multimedia
- Vol. 28 (15209210) , 1689-1703
TransGOP-R: Transformer-Based Real-World Gaze Object Prediction
The goal of gaze object prediction (GOP) is to predict human gaze objects and categories. However, existing methods require additional head priors or filter the results before evaluation, which is an obstacle for real-world applications. To this end, this paper proposes a Transformer-based Gaze Object Prediction under Real-world setting (TransGOP-R), which does not rely on any head prior input and evaluates end-to-end. We first design a head location module to generate human head location information from a head query. Then, an error analysis demonstrates that the primary error source of the existing GOP model is in gaze estimation, which is caused by the difficulty in predicting gaze points by directly regressing heatmaps. Therefore, we introduce cone prediction into the model training stage, allowing the middle-layer features of the gaze regressor to build the relationship between the target human and objects before regressing the gaze point. An oriented gradient mechanism is proposed in this process to ensure the object detection performance is not affected by cone information. Finally, we conducted very detailed and sufficient experiments to verify the superiority of our method on the GOO-Synth and GOO-Real datasets. At the same time, we also achieve advantages compared to the human-target gaze estimation methods on the GazeFollowing, VideoAttentionTarget, and ChildPlay datasets.
- 20 November 2025
- Published by Elsevier in Pattern Recognition
- Vol. 172
Unlocking human intent perception through multimodal large models
- 20 November 2025
- Published by Elsevier in Pattern Recognition
- Vol. 172
Unlocking human intent perception through multimodal large models