Chao Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Chao Xu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Chong Ling
Army Academy of Artillery and Air Defense of PLA, Hefei, China
1 shared publication - Hao Li
Army Engineering University of PLA, Nanjing, China
1 shared publication - Jun Li
China University of Geosciences
1 shared publication - Xue Song
Fudan University
1 shared publication - Yongfeng Liu
Yangzhou University
1 shared publication - Zhan Yuan
Unit of PLA
1 shared publication
Latest Publications
- 14 November 2023
- Published by Springer Nature
- p. 78-90
Amortized Variational Inference via Nosé-Hoover Thermostat Hamiltonian Monte Carlo
- 14 November 2023
- Published by Springer Nature
- p. 78-90
Amortized Variational Inference via Nosé-Hoover Thermostat Hamiltonian Monte Carlo
- Open Access
- 7 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 126634-126642
Siamese Contour Segmentation Network for Multi-State Object Tracking
Tracking multiple states of the target simultaneously is currently a research hotspot in object tracking. Existing methods obtain the initial bounding box with multi-scale search, anchor-based regression, or anchor-free regression. Then, each pixel within bounding box is classified and the mask of target is used to fit a fine bounding box for accurate tracking. However, their computation complexity is high and the accuracy of mask and fine bounding box is limited by initial bounding box. Based on SiamMask, the Siamese contour segmentation network (SiamCS) is proposed for multi-state object tracking to address these issues. This end-to-end network eliminates need to pre-define anchor based on prior knowledge, and reduces hyperparameters. With SiamCS, multi-state object tracking method formulates object tracking as region classification and contour regression to obtain bounding box, contour, and mask of target at the same time. Moreover, according to geometric meaning of definite integral and difficulty of sample, difficulty sensitive contour-intersection over union loss function is proposed to solve the problem of independent regression of contour parameters. Extensive experiments on OTB100 (92.5%Precision), UAV123 (83.5%Precision), LaSOT (64.8%AUC), TrackingNet (81.5%AUC), GOT-10K (66.3%AO), and VOT2020 (53.5%EAO) show that SiamCS outperforms many state-of-the-art trackers and achieves leading performance.
- 7 November 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 126634-126642
Open AccessSiamese Contour Segmentation Network for Multi-State Object Tracking
Tracking multiple states of the target simultaneously is currently a research hotspot in object tracking. Existing methods obtain the initial bounding box with multi-scale search, anchor-based regression, or anchor-free regression. Then, each pixel within bounding box is classified and the mask of target is used to fit a fine bounding box for accurate tracking. However, their computation complexity is high and the accuracy of mask and fine bounding box is limited by initial bounding box. Based on SiamMask, the Siamese contour segmentation network (SiamCS) is proposed for multi-state object tracking to address these issues. This end-to-end network eliminates need to pre-define anchor based on prior knowledge, and reduces hyperparameters. With SiamCS, multi-state object tracking method formulates object tracking as region classification and contour regression to obtain bounding box, contour, and mask of target at the same time. Moreover, according to geometric meaning of definite integral and difficulty of sample, difficulty sensitive contour-intersection over union loss function is proposed to solve the problem of independent regression of contour parameters. Extensive experiments on OTB100 (92.5%Precision), UAV123 (83.5%Precision), LaSOT (64.8%AUC), TrackingNet (81.5%AUC), GOT-10K (66.3%AO), and VOT2020 (53.5%EAO) show that SiamCS outperforms many state-of-the-art trackers and achieves leading performance.
- 28 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 19-23
Object Detection in Visible and Infrared missile borne fusion image
At present, common object detection methods for missile borne images often perform poorly. Because the experimental data of missile borne images is difficult to obtain, the target is small, and the imaging environment is complex and random, it is a challenge to build an appropriate object detection model for such missile borne images. Based on the classic YOLOv5, the paper constructs an aerial platform image target detection model YOLO v5mb, which is suitable for missile borne images. The model can accurately detect targets in single-mode visible or infrared missile borne images. In addition, the fusion layer architecture in YOLO v5mb makes it suitable for multimode visible and infrared missile borne fusion images object detection.
- 28 October 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 19-23
Object Detection in Visible and Infrared missile borne fusion image
At present, common object detection methods for missile borne images often perform poorly. Because the experimental data of missile borne images is difficult to obtain, the target is small, and the imaging environment is complex and random, it is a challenge to build an appropriate object detection model for such missile borne images. Based on the classic YOLOv5, the paper constructs an aerial platform image target detection model YOLO v5mb, which is suitable for missile borne images. The model can accurately detect targets in single-mode visible or infrared missile borne images. In addition, the fusion layer architecture in YOLO v5mb makes it suitable for multimode visible and infrared missile borne fusion images object detection.