Xu Tao (徐 涛)

Xu Tao (徐 涛)

Ph.D, Associate Prof.

Xu Tao (徐 涛 ) is associate professor with the School of Information Science and Engineering, University of Jinan(济南大学). He received the M.S. degree in computer science from China University of Petroleum(中国石油大学), and the Ph.D. degree in computer science from IRIP Lab, Beihang University(北京航空航天大学).

Research interest includes remote sensing image analysis and human-computer interaction, especially interested in the following topics:

  • Embodied Intelligence and Human-Robot Interaction.
  • Detection, Segmentation and Change Detection from Remote Sensing images.

🔥 News

  • 2026.3.21:  Our paper “FSFENet: An Efficient Frequency-Spatial Feature Extraction Network for Remote Sensing Change Detection” has been accepted by IJCNN 2026(CCF-C).
  • 2026.1.18:  Our paper “Multimodal Fusion-based IPCLIP Network for Mixed Reality Surgical Assistance” has been accepted by ICASSP 2026(CCF-B).

📝 Publications

IEEE JSTARS
sym

HFIFNet: Hierarchical Feature Interaction Network with Multi-scale Fusion for Change Detection

Mingzhi Han, Tao Xu,Qingjie Liu, Xiaohui Yang, Jing Wang, and Jiaqi Kong

  • We propose a HFIFM, which achieves optimized multichannel feature interaction and enhances the distinguishability between weak targets, background, and unchanged targets. Simultaneously, the module also enhances the representation and compensates for the loss of spatial feature of changed targets at a single scale by integrating cross scale feature fusion.
  • We employ a VMB and propose a SFLM to enhance thesaliency of spatial features extracted by the VMB. The distinguishability between weak targets and irrelevant spatial features is further improved through the global feature perception capability of the VMB.
  • Comprehensive experiments on the LEVIR-CD, S2Looking and NALand datasets demonstrate the effectiveness of our method for CD tasks, achieving state-of-the-art (SOTA) performance on most metrics.
  • 2026
    1. Jiahui Sun, Tao Xu, Xiaohui Yang, Tongzhen Si, Xiaoli Liu. Multimodal Fusion-based IPCLIP Network for Mixed Reality Surgical Assistance[C]. ICASSP 2026, Accepted.
    2. Hao Zhang, Tao Xu, et al.FSFENet: An Efficient Frequency-Spatial Feature Extraction Network for Remote Sensing Change Detection[C].IJCNN 2026, Accepted.
  • 2025
    1. Han M, Xu T, Liu Q, et al. HFIFNet: Hierarchical Feature Interaction Network with Multi-scale Fusion for Change Detection[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing(J-STARS),2025, 18:4318-4330. (SCI) code
    2. Han Zhang, Tao Xu, Xiaohui Yang, Yuqian Zhang, Chenxin Xiao. Group Convolution Attention based Multi-Scale Object Detection of Remote Sensing Images[C]. 2025 IEEE International Geoscience and Remote Sensing Symposium（IGARSS 2025）, Brisbane, Australia, 2025: 6031-6035.
    3. Chenxin Xiao, Tao Xu, Xiaohui Yang, Wenjun Kang, Junyuan Zang. EANV: An Edge Attention Network with VMamba Block for Semantic Segmentation from Remote Sensing Images[C]. 2025 IEEE International Geoscience and Remote Sensing Symposium（IGARSS 2025）, Brisbane, Australia, 2025: 2131-2135.
    4. Junyuan Zang , Tao Xu, Xiaohui Yang , Yuqian Zhang and Wenjun Kang. MSD-SEG: Multi-scale Deformable Convolution Network for Segmentation of Remote Sensing Images[C]. International Conference on Intelligent Computing(ICIC2025). Singapore: Springer Nature Singapore, 2025: 278-290.
    5. Yijing Zhai, Tao Xu, Yuqian Zhang, Baozhu Wan. Curriculum-Learned Masked Pretraining Models for Remote Sensing Building Detection[C]. International Conference on Intelligent Computing(ICIC2025). Singapore: Springer Nature Singapore, 2025: 304-315.
  • 2024
    1. Han M, Xu T, Liu Q, et al. MAMI-CD: Multi-stage Attention Network for Change Detection with Mixed Feature Interaction[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing(J-STARS),2024, 17:14908-14918. (SCI) code
    2. Q. Lu, T. Xu, J. Dong, Q. Liu and X. Yang. Improving Object Detection from Remote Sensing Images via Self-supervised Adaptive Fusion Networks[J]. IEEE Geoscience and Remote Sensing Letters(GRSL), 2024, 21:1-5. (SCI)
    3. Jiahui Sun，Tao Xu，Zhaoyu Li, et al. ARHands: A Hand Gesture Dataset with HCI in Augmented Reality Surgery[C]. 2024 6th International Conference on Frontier Technologies of Information and Computer (ICFTIC), Qingdao, China, 2024:914-917.
  • 2023
    1. Huang Z, Liu Q, Zhou H, et al. Building Detection From Panchromatic and Multispectral Images With Dual-Stream Asymmetric Fusion Networks[J]. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing(J-STARS), 2023, 16: 3364-3377. (SCI)
    2. Yu K, Zhang Y, Hou B, et al. Large Window Attention Based Transformer Network for Change Detection of Remote Sensing Images[C]. International Conference on Image and Graphics(ICIG). Cham: Springer Nature Switzerland, 2023: 308-319. (EI)
    3. Han M, Liu L, Xu T, et al. Non-Agricultural Change Detection in Multi-Scale and Multi-Season Remote Sensing Images[C]. 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2023: 1-5. (EI)
    4. Zang J, Zhang Y, Xu T, et al. Introduction and Analysis of M-TSLand: A Multi-Temporal and Multi-Scale Remote Sensing Farmland Dataset[C]. Proceedings of the 2023 6th International Conference on Big Data Technologies(ICBDT). 2023: 151-157. (EI)
    5. Xiao C, Liu L, Xie G, et al. A large dataset with CNN-based segmentation analysis for land cover from large river basins[C]. 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2023: 1-5. (EI)
    6. Zhang H, Bi M, Xu T, et al. A multi-scale dataset with comprehensive analysis for building detection from remote sensing images[C]. 2023 16th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2023: 1-5. (EI)
    7. Li M, Xu T, Chen X, et al. Design of 120GHZ FMCW Radar Liquid Level Measurement System[C]. 2023 5th International Conference on Intelligent Control, Measurement and Signal Processing (ICMSP). IEEE, 2023: 755-758. (EI)
    8. Li M, Xu T, Chen X, et al. Research on algorithm of FMCW radar level gauge[C]. 2023 5th International Conference on Electronics and Communication, Network and Computer Technology (ECNCT). IEEE, 2023: 270-274. (EI)
    9. SE-UNet: Channel Attention based UNet for Water body Segmentation from SAR Image[C]
    10. Z. Li,T. Xu, X. Yang, J. Sun and G. Zhu, Variable Scale Gesture Recognition: A Dataset and Comprehensive Analysis[C]. 2023 3rd International Conference on Robotics, Automation and Intelligent Control (ICRAIC), Zhangjiajie, China, 2023, pp. 110-114.
  • 2022
    1. Li W, Xu T, Wang Y, et al. UJN-SAR: A Large Dataset with Experimental Analysis for Water Body Segmentation from SAR Images[C]. International Conference on Image, Vision and Intelligent Systems. Singapore: Springer Nature Singapore, 2022: 227-238. (EI)
    2. Yu K,Xu T, Hou B, et al. UJN-CD: A Large Scale Date set for Change Detection with Comprehensive Analysis Based on Deep Learning[C]. International Conference on Image, Vision and Intelligent Systems. Singapore: Springer Nature Singapore, 2022: 432-441.
    3. Lu Q, Dong W J, Dong Y, et al. EMN: An Edge Mask based Network for Building Detection from Remote Sensing Images[C]. Proceedings of the 5th International Conference on Big Data Technologies. 2022: 242-247. (EI)
    4. Li Y, Han S Y, Xu T, et al. Cascade Detector Based on Multi-scale Features for Small Objects in Various Backgrounds of Remote Sensing[C]. International Conference on Image, Vision and Intelligent Systems. Singapore: Springer Nature Singapore, 2022: 310-319. (EI)
    5. Shen Y, Xu T, Liu L Y, et al. Laplacian Filter Embedding based Attention Network for Land Segmentation from Remote Sensing Images[C]. Proceedings of the 5th International Conference on Big Data Technologies. 2022: 284-290. (EI)
  • 2021
    1. Yin H, Zhang C, Han Y, et al. Improved semantic segmentation method using edge features for winter wheat spatial distribution extraction from Gaofen-2 images[J]. Journal of Applied Remote Sensing, 2021, 15(2): 028501-028501.(SCI)
    2. Qiu X, Feng Z, Xu T, et al. Research on Intention Flexible Map Algorithm for Elderly Escort Robot[J]. Scientific Programming, 2021, 2021: 1-14.(SCI)
    3. Li Y, Han S Y, Shen Y, et al. UJN-traffic: a benchmark dataset for performance evaluation of traffic element classification[C]. Intelligent Equipment, Robots, and Vehicles: 7th International Conference on Life System Modeling and Simulation, LSMS 2021 and 7th International Conference on Intelligent Computing for Sustainable Energy and Environment, ICSEE 2021, Hangzhou, China, October 22–24, 2021, Proceedings, Part III 7. Springer Singapore, 2021: 184-193.(EI)
    4. Shen Y, Wang Y, Yang S, et al. UJN-Land: A Large-Scale High-Resolution Parcel of Land of Multi-temporal Dataset with CNN Based Semantic Segmentation Analysis[C]. Intelligent Equipment, Robots, and Vehicles: 7th International Conference on Life System Modeling and Simulation, LSMS 2021 and 7th International Conference on Intelligent Computing for Sustainable Energy and Environment, ICSEE 2021, Hangzhou, China, October 22–24, 2021, Proceedings, Part III 7. Springer Singapore, 2021: 204-213. (EI)
    5. Yu P, Yang S, Shen Y, et al. Design and Implementation of Security Labeling System Based on Neighborhood Rough Set[C]//Intelligent Equipment, Robots, and Vehicles: 7th International Conference on Life System Modeling and Simulation, LSMS 2021 and 7th International Conference on Intelligent Computing for Sustainable Energy and Environment, ICSEE 2021, Hangzhou, China, October 22–24, 2021, Proceedings, Part III 7. Springer Singapore, 2021: 462-471. (EI)
  • 2020
    1. Lu Z, Liu K, Zhang Y, et al. Building Detection via Complementary Convolutional Features of Remote Sensing Images[C]. Pattern Recognition and Computer Vision: Third Chinese Conference(PRCV), Nanjing, China, October 16–18, 2020, Proceedings, Part I 3. Springer International Publishing, 2020: 638-647. (EI，CCF-C)
    2. Wang W, Zhang X, Dong L, et al. Network attack detection based on domain attack behavior analysis[C]//2020 13th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2020: 962-965.
    3. Chen J, Cao A, Zhang Y, et al. Tumor-assisted diagnosis based on U-Net network[C]//2020 13th International Congress on Image and Signal Processing, BioMedical Engineering and Informatics (CISP-BMEI). IEEE, 2020: 739-742.
  • Other papers
    1. Lu Z, Xu T, Liu K, et al. 5M-Building: A large-scale high-resolution building dataset with CNN based detection analysis[C]. 2019 IEEE 31st International Conference on Tools with Artificial Intelligence (ICTAI). IEEE, 2019: 1385-1389. (EI，CCF-C)
    2. Yu P, Xu T, Zheng B, et al. Research and implementation of multiscale dynamic terrain[C]. 2018 International Conference on Security, Pattern Analysis, and Cybernetics (SPAC). IEEE, 2018: 217-222.(Best Paper Award)
    3. Feng Z, Yang B, Xu T, et al. FM: Flexible maping from one gesture to multiple semantics[J]. Information Sciences, 2018, 467: 654-669.(SCI)
    4. Han R, Feng Z, Xu T, et al. Multi-sensors based 3D gesture recognition and interaction in virtual block game[C]. 2017 International Conference on Virtual Reality and Visualization (ICVRV). IEEE, 2017: 391-392. (EI，CCF-C)
    5. Xu T, Feng Z, Dong L, et al. 2D Human Parsing with Deep Skin Model and Part-Based Model Inference[C]. Intelligent Computing Theories and Application: 13th International Conference, ICIC 2017, Liverpool, UK, August 7-10, 2017, Proceedings, Part II 13. Springer International Publishing, 2017: 776-787. (EI，CCF-C)
    6. Xu T, Zhang Z, Wang Y. Patch-wise skin segmentation of human body parts via deep neural networks[J]. Journal of Electronic Imaging, 2015, 24(4): 043009-043009.(SCI)
    7. Feng Z, Xu T, Lv N, et al. Behavioral Model Tracking of Hand Gestures[C]. 2015 International Conference on Virtual Reality and Visualization (ICVRV). IEEE, 2015: 101-108.
    8. Xu T, Wang Y, Zhang Z. Pixel‐wise skin colour detection based on flexible neural tree[J]. IET Image Processing, 2013, 7(8): 751-761.(SCI)
    9. Xu T, Wang Y, Zhang Z. Towards independent color space selection for human skin detection[C]. Advances in Multimedia Information Processing–PCM 2012: 13th Pacific-Rim Conference on Multimedia, Singapore, December 4-6, 2012. Proceedings 13. Springer Berlin Heidelberg, 2012: 337-346. (EI)
    10. Xu T, Wang C, Wang Y, et al. Saliency model based head pose estimation by sparse optical flow[C]. The First Asian Conference on Pattern Recognition(ACPR). IEEE, 2011: 575-579. (EI)

📖 Projects

sym

Intelligent interpretation of remote sensing images：Remote sensing images have the advantages of wide observation range, high observation frequency, and objective and realistic observation data. Obtaining useful information from remote sensing images has become an effective means for current social development and governance. The intelligent interpretation of remote sensing images mainly includes research topics such as object detection, object segmentation, change detection, and object and scene classification. The research group has developed research directions such as building detection, water body segmentation, land use classification, and change detection based on actual needs.

sym

Human-computer interaction：Human computer interaction is one of the key technologies for robots to assist human life and work. The laboratory has advanced equipment such as mobile robots and robotic arms, and conducts multimodal human-computer interaction research for elderly care and virtual experiments, with a focus on gesture recognition, multimodal fusion, and understanding of interaction intentions.

🎖 Members

  • Faculty : Feng Zhiquan(PI), Xu Tao, Yang Xiaohui, Si Tongzhen, Guo Qingbei, Zhang Feng.
  • Master Candidates:
    1. Xiao Chenxin, Zhang Han, Han Mingzhi, LiZhaoyu, Zang Junyuan;
    2. Zhu Guangze, Sun Jiahui, Zhai Yijing;
    3. Tian Qizong, Zhang Hao, Zhiyu Guo.

💬 Invited Talks

  • 2025.11.22, Brain-Like Computing and Artificial Intelligence Applications（类脑计算与人工智能应用）, MShandong Neuroscience Society Yihe Forum（山东省神经科学学会沂河论坛）.
  • 2022.11.23, Research on the Application and Intelligent Interpretation Technology of High Resolution Satellites in Shandong Province, Mount Taishan Meteorological Forum.

    💻 Datasets

sym

UJN-3M-HANDS：Large-scale dataset is necessary for training of deep neural network-based hand gesture classification models. According to the requirements of human-interaction in virtual reality-oriented teaching environment, a large-scale depth-based hand gesture dataset was proposed. The dataset, named UJN-3M-HANDS, contained more than 3 million depth gesture samples. There are 17 static gestures and 34 dynamic gestures in UJN-3M-HANDS. Some static gesture samples were demonstrated in Figure. Compared with other data sets, UJN-3M-HANDS proposed is designed for specific teaching application scenarios, and the categories of gestures are consistent with those used in experimental teaching. In the stage of gesture collection, the distance between the camera and the collector is set to more than 2 meters. The advantage of UJN-3M-HANDS is that the gesture samples were acquired from 2000 individuals, while there are many replication samples in other datasets.

sym

5M-Buildings：Buildings dataset is constructed via manually annotation based on GaoFen satellite, which images are with high spatial resolution. In our work, the building categories are divided into house buildings, factory buildings and other buildings, etc. Compared with existing annotation methods, the proposed annotation process is more efficient and accurate, because we annotate panchromatic images with high spatial resolution, and then fuse panchromatic and multispectral images. The traditional annotation process is divided into two parallel sub-processes, i.e., fast panchromatic image annotation and multi-spectral image fusion. Buildings now consists of three distinct categories building categories, i.e., house buildings, factory buildings and other buildings. The number of categories continues to increase. The dataset contains more than 10,000 images. The spatial resolution of sample images is less than 1 meter, which covering various types of buildings on the mainland. Images under various conditions of different weather, seasons and imaging conditions are selected for buildings annotation. Sample images of Buildings have the characteristics of large-scale, multi-scale, intra-class diversity and inter-class similarity in terms of building. Although existing remote sensing datasets contains more scenarios, the number of samples cannot meet the needs of model training for CNN based target detection or classification. To the best of our knowledge, the Buildings dataset proposed in this paper is the largest high-resolution remote sensing dataset for building detection.