Xiang Liu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Xiang Liu has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Hongyuan Wang
School of Astronautics, Harbin Institute of Technology, Harbin 150001, China
8 shared publications - Qianhao Ning
Space Optical Engineering Research Center, Harbin Institute of Technology, Harbin 150001, China
5 shared publications - Zhiqiang Yan
Research Center for Space Optical Engineering, Harbin Institute of Technology, Harbin 150001, China
5 shared publications Weichun Chen
China Acad Space Technol, Qian Xuesen Lab Space Technol, Beijing 100080, Peoples R China
3 shared publicationsXinlong Chen
China Acad Space Technol, Qian Xuesen Lab Space Technol, Beijing 100080, Peoples R China
3 shared publications- Bingwen Wang
Research Center for Space Optical Engineering, School of Aeronautics, Harbin Institute of Technology, Harbin, China
2 shared publications
Latest Publications
- Open Access
- 9 March 2024
- Published by Elsevier in Chinese Journal of Aeronautics
- Vol. 37 (7) , 439-449
Filtering and regret network for spacecraft component segmentation based on gray images and depth maps
- 9 March 2024
- Published by Elsevier in Chinese Journal of Aeronautics
- Vol. 37 (7) , 439-449
Open AccessFiltering and regret network for spacecraft component segmentation based on gray images and depth maps
- Open Access
- 30 September 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (19) , 4786
Exploiting Temporal–Spatial Feature Correlations for Sequential Spacecraft Depth Completion
The recently proposed spacecraft three-dimensional (3D) structure recovery method based on optical images and LIDAR has enhanced the working distance of a spacecraft’s 3D perception system. However, the existing methods ignore the richness of temporal features and fail to capture the temporal coherence of consecutive frames. This paper proposes a sequential spacecraft depth completion network (S2DCNet) for generating accurate and temporally consistent depth prediction results, and it can fully exploit temporal–spatial coherence in sequential frames. Specifically, two parallel convolution neural network (CNN) branches were first adopted to extract the features latent in different inputs. The gray image features and the depth features were hierarchically encapsulated into unified feature representations through fusion modules. In the decoding stage, the convolutional long short-term memory (ConvLSTM) networks were embedded with the multi-scale scheme to capture the feature spatial–temporal distribution variation, which could reflect the past state and generate more accurate and temporally consistent depth maps. In addition, a large-scale dataset was constructed, and the experiments revealed the outstanding performance of the proposed S2DCNet, achieving a mean absolute error of 0.192 m within the region of interest.
- 30 September 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (19) , 4786
Open AccessExploiting Temporal–Spatial Feature Correlations for Sequential Spacecraft Depth Completion
The recently proposed spacecraft three-dimensional (3D) structure recovery method based on optical images and LIDAR has enhanced the working distance of a spacecraft’s 3D perception system. However, the existing methods ignore the richness of temporal features and fail to capture the temporal coherence of consecutive frames. This paper proposes a sequential spacecraft depth completion network (S2DCNet) for generating accurate and temporally consistent depth prediction results, and it can fully exploit temporal–spatial coherence in sequential frames. Specifically, two parallel convolution neural network (CNN) branches were first adopted to extract the features latent in different inputs. The gray image features and the depth features were hierarchically encapsulated into unified feature representations through fusion modules. In the decoding stage, the convolutional long short-term memory (ConvLSTM) networks were embedded with the multi-scale scheme to capture the feature spatial–temporal distribution variation, which could reflect the past state and generate more accurate and temporally consistent depth maps. In addition, a large-scale dataset was constructed, and the experiments revealed the outstanding performance of the proposed S2DCNet, achieving a mean absolute error of 0.192 m within the region of interest.
- Open Access
- 13 August 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (16) , 4010
Space-Based THz Radar Fly-Around Imaging Simulation for Space Targets Based on Improved Path Tracing
Aiming at the space target detection application of a space-based terahertz (THz) radar, according to the imaging mechanism of broadband THz radars, a THz radar imaging simulation method based on improved path tracing is proposed. Firstly, the characterization method of THz scattering characteristics based on Kirchhoff’s approximation method is introduced. The multi-parameter THz bidirectional reflectance distribution function (THz-BRDF) models of aluminum (Al), white-painted Al, and polyimide film at 0.215 THz are fitted according to the theoretical data, with fitting errors below 4%. Then, the THz radar imaging simulation method based on improved path tracing is presented in detail. The simulation method utilizes path tracing to simulate parallelized THz radar echo signal data, considering multi-path energy scattering based on the THz-BRDF model. Finally, we conducted THz radar imaging simulation experiments. The influences in the imaging process of different fly-around motions are analyzed, and a comparison experiment is conducted with the fast-physical optics (FPO) method. The comparative results indicate that the proposed method exhibits richer and more realistic features compared with the FPO method. The simulation experiments results demonstrate that the proposed method can provide a data source for ground algorithm testing of THz radars, particularly in evaluating the target detection and recognition algorithm based on deep learning, providing strong support for the application of space-based THz radars in the future.
- 13 August 2023
- Published by MDPI AG in Remote Sensing
- Vol. 15 (16) , 4010
Open AccessSpace-Based THz Radar Fly-Around Imaging Simulation for Space Targets Based on Improved Path Tracing
Aiming at the space target detection application of a space-based terahertz (THz) radar, according to the imaging mechanism of broadband THz radars, a THz radar imaging simulation method based on improved path tracing is proposed. Firstly, the characterization method of THz scattering characteristics based on Kirchhoff’s approximation method is introduced. The multi-parameter THz bidirectional reflectance distribution function (THz-BRDF) models of aluminum (Al), white-painted Al, and polyimide film at 0.215 THz are fitted according to the theoretical data, with fitting errors below 4%. Then, the THz radar imaging simulation method based on improved path tracing is presented in detail. The simulation method utilizes path tracing to simulate parallelized THz radar echo signal data, considering multi-path energy scattering based on the THz-BRDF model. Finally, we conducted THz radar imaging simulation experiments. The influences in the imaging process of different fly-around motions are analyzed, and a comparison experiment is conducted with the fast-physical optics (FPO) method. The comparative results indicate that the proposed method exhibits richer and more realistic features compared with the FPO method. The simulation experiments results demonstrate that the proposed method can provide a data source for ground algorithm testing of THz radars, particularly in evaluating the target detection and recognition algorithm based on deep learning, providing strong support for the application of space-based THz radars in the future.
- 15 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Aerospace and Electronic Systems
- Vol. PP (00189251) , 1-13
Spacecraft depth completion based on the gray image and the sparse depth map
Perceiving the three-dimensional (3D) structure of the spacecraft is a prerequisite for successfully executing many on-orbit space missions, and it can provide critical input for many downstream vision algorithms. In this paper, we propose to sense the 3D structure of spacecraft using light detection and ranging sensor (LIDAR) and a monocular camera. To this end, Spacecraft Depth Completion Network (SDCNet) is proposed to recover the dense depth map based on gray image and sparse depth map. Specifically, SDCNet decomposes the object-level spacecraft depth completion task into foreground segmentation subtask and foreground depth completion subtask, which segments the spacecraft region first and then performs depth completion on the segmented foreground area. In this way, the background interference to foreground spacecraft depth completion is effectively avoided. Moreover, an attention-based feature fusion module is also proposed to aggregate the complementary information between different inputs, which deduces the correlation between different features along the channel and the spatial dimension sequentially. Besides, four metrics are also proposed to evaluate object-level depth completion performance, which can more intuitively reflect the quality of spacecraft depth completion results. Finally, a large-scale satellite depth completion dataset is constructed for training and testing spacecraft depth completion algorithms. Empirical experiments on the dataset demonstrate the effectiveness of the proposed SDCNet, which achieves 0.225m mean absolute error of interest and 0.778m mean absolute truncation error, surpassing state-of-the-art methods by a large margin. The spacecraft pose estimation experiment is also conducted based on the depth completion results, and the experimental results indicate that the predicted dense depth map could meet the needs of downstream vision tasks.
- 15 June 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Aerospace and Electronic Systems
- Vol. PP (00189251) , 1-13
Spacecraft depth completion based on the gray image and the sparse depth map
Perceiving the three-dimensional (3D) structure of the spacecraft is a prerequisite for successfully executing many on-orbit space missions, and it can provide critical input for many downstream vision algorithms. In this paper, we propose to sense the 3D structure of spacecraft using light detection and ranging sensor (LIDAR) and a monocular camera. To this end, Spacecraft Depth Completion Network (SDCNet) is proposed to recover the dense depth map based on gray image and sparse depth map. Specifically, SDCNet decomposes the object-level spacecraft depth completion task into foreground segmentation subtask and foreground depth completion subtask, which segments the spacecraft region first and then performs depth completion on the segmented foreground area. In this way, the background interference to foreground spacecraft depth completion is effectively avoided. Moreover, an attention-based feature fusion module is also proposed to aggregate the complementary information between different inputs, which deduces the correlation between different features along the channel and the spatial dimension sequentially. Besides, four metrics are also proposed to evaluate object-level depth completion performance, which can more intuitively reflect the quality of spacecraft depth completion results. Finally, a large-scale satellite depth completion dataset is constructed for training and testing spacecraft depth completion algorithms. Empirical experiments on the dataset demonstrate the effectiveness of the proposed SDCNet, which achieves 0.225m mean absolute error of interest and 0.778m mean absolute truncation error, surpassing state-of-the-art methods by a large margin. The spacecraft pose estimation experiment is also conducted based on the depth completion results, and the experimental results indicate that the predicted dense depth map could meet the needs of downstream vision tasks.
- Open Access
Binary Feature Description of 3D Point Cloud Based on Retina-like Sampling on Projection Planes
A binary feature description and registration algorithm for a 3D point cloud based on retina-like sampling on projection planes (RSPP) are proposed in this paper. The algorithm first projects the point cloud within the support radius around the key point to the XY, YZ, and XZ planes of the Local Reference Frame (LRF) and performs retina-like sampling on the projection plane. Then, the binarized Gaussian density weight values at the sampling points are calculated and encoded to obtain the RSPP descriptor. Finally, rough registration of point clouds is performed based on the RSPP descriptor, and the RANSAC algorithm is used to optimize the registration results. The performance of the proposed algorithm is tested on public point cloud datasets. The test results show that the RSPP-based point cloud registration algorithm has a good registration effect under no noise, 0.25 mr, and 0.5 mr Gaussian noise. The experimental results verify the correctness and robustness of the proposed registration method, which can provide theoretical and technical support for the 3D point cloud registration application.
Open AccessBinary Feature Description of 3D Point Cloud Based on Retina-like Sampling on Projection Planes
A binary feature description and registration algorithm for a 3D point cloud based on retina-like sampling on projection planes (RSPP) are proposed in this paper. The algorithm first projects the point cloud within the support radius around the key point to the XY, YZ, and XZ planes of the Local Reference Frame (LRF) and performs retina-like sampling on the projection plane. Then, the binarized Gaussian density weight values at the sampling points are calculated and encoded to obtain the RSPP descriptor. Finally, rough registration of point clouds is performed based on the RSPP descriptor, and the RANSAC algorithm is used to optimize the registration results. The performance of the proposed algorithm is tested on public point cloud datasets. The test results show that the RSPP-based point cloud registration algorithm has a good registration effect under no noise, 0.25 mr, and 0.5 mr Gaussian noise. The experimental results verify the correctness and robustness of the proposed registration method, which can provide theoretical and technical support for the 3D point cloud registration application.
- 16 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Aerospace and Electronic Systems
- Vol. 59 (1) , 507-518
Position Awareness Network for Noncooperative Spacecraft Pose Estimation Based on Point Cloud
Spacecraft pose estimation plays a vital role in many on-orbit space missions, such as rendezvous and docking, debris removal, and on-orbit maintenance. At present, the mainstream descriptors-based pose estimation methods ignore the fact that satellite point cloud contains many similar structures, generating numerous mismatched correspondence pairs and leading to low pose estimation accuracy. This paper proposes a Position Awareness Network (PANet) for spacecraft pose estimation. Specifically, the point cloud is first fed into a hierarchical embedding network to extract the key points and construct local structural descriptors. We also build the relative position features by encoding the relative position between key points and reference points. The matching matrix between point clouds is then calculated by comprehensively considering the local structure descriptors and relative location features. In this way, the problem of ambiguous matching caused by similar local structures is avoided. Finally, weighted singular value decomposition (SVD) is utilized to solve the pose between the point clouds based on the correspondence pairs generated by the matching matrix. Besides, a large-scale satellite point cloud dataset is also constructed for training and testing pose estimation algorithms. Empirical experiments on the dataset demonstrate the effectiveness of the proposed PANet, which achieves 1.18° rotation error and 0.136m translate error, surpassing state-of-the-art methods by a large margin.
- 16 June 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Aerospace and Electronic Systems
- Vol. 59 (1) , 507-518
Position Awareness Network for Noncooperative Spacecraft Pose Estimation Based on Point Cloud
Spacecraft pose estimation plays a vital role in many on-orbit space missions, such as rendezvous and docking, debris removal, and on-orbit maintenance. At present, the mainstream descriptors-based pose estimation methods ignore the fact that satellite point cloud contains many similar structures, generating numerous mismatched correspondence pairs and leading to low pose estimation accuracy. This paper proposes a Position Awareness Network (PANet) for spacecraft pose estimation. Specifically, the point cloud is first fed into a hierarchical embedding network to extract the key points and construct local structural descriptors. We also build the relative position features by encoding the relative position between key points and reference points. The matching matrix between point clouds is then calculated by comprehensively considering the local structure descriptors and relative location features. In this way, the problem of ambiguous matching caused by similar local structures is avoided. Finally, weighted singular value decomposition (SVD) is utilized to solve the pose between the point clouds based on the correspondence pairs generated by the matching matrix. Besides, a large-scale satellite point cloud dataset is also constructed for training and testing pose estimation algorithms. Empirical experiments on the dataset demonstrate the effectiveness of the proposed PANet, which achieves 1.18° rotation error and 0.136m translate error, surpassing state-of-the-art methods by a large margin.
- Open Access
- 15 June 2022
- Published by MDPI AG in Remote Sensing
- Vol. 14 (12) , 2868
Physics-Based TOF Imaging Simulation for Space Targets Based on Improved Path Tracing
Aiming at the application of close-up space measurement based on time-of-flight (TOF) cameras, according to the analysis of the characteristics of the space background environment and the imaging characteristics of the TOF camera, a physics-based amplitude modulated continuous wave (AMCW) TOF camera imaging simulation method for space targets based on the improved path tracing is proposed. Firstly, the microfacet bidirectional reflection distribution function (BRDF) model of several typical space target surface materials is fitted according to the measured BRDF data in the TOF camera response band to make it physics-based. Secondly, an improved path tracing algorithm is developed to adapt to the TOF camera by introducing a cosine component to characterize the modulated light in the TOF camera. Then, the imaging link simulation model considering the coupling effects of the BRDF of materials, the suppression of background illumination (SBI), optical system, detector, electronic equipment, platform vibration, and noise is established, and the simulation images of the TOF camera are obtained. Finally, ground tests are carried out, and the test shows that the relative error of the grey mean, grey variance, depth mean, and depth variance is 2.59%, 3.80%, 18.29%, and 14.58%, respectively; the MSE, SSIM, and PSNR results of our method are also better than those of the reference method. The ground test results verify the correctness of the proposed simulation model, which can provide image data support for the ground test of TOF camera algorithms for space targets.
- 15 June 2022
- Published by MDPI AG in Remote Sensing
- Vol. 14 (12) , 2868
Open AccessPhysics-Based TOF Imaging Simulation for Space Targets Based on Improved Path Tracing
Aiming at the application of close-up space measurement based on time-of-flight (TOF) cameras, according to the analysis of the characteristics of the space background environment and the imaging characteristics of the TOF camera, a physics-based amplitude modulated continuous wave (AMCW) TOF camera imaging simulation method for space targets based on the improved path tracing is proposed. Firstly, the microfacet bidirectional reflection distribution function (BRDF) model of several typical space target surface materials is fitted according to the measured BRDF data in the TOF camera response band to make it physics-based. Secondly, an improved path tracing algorithm is developed to adapt to the TOF camera by introducing a cosine component to characterize the modulated light in the TOF camera. Then, the imaging link simulation model considering the coupling effects of the BRDF of materials, the suppression of background illumination (SBI), optical system, detector, electronic equipment, platform vibration, and noise is established, and the simulation images of the TOF camera are obtained. Finally, ground tests are carried out, and the test shows that the relative error of the grey mean, grey variance, depth mean, and depth variance is 2.59%, 3.80%, 18.29%, and 14.58%, respectively; the MSE, SSIM, and PSNR results of our method are also better than those of the reference method. The ground test results verify the correctness of the proposed simulation model, which can provide image data support for the ground test of TOF camera algorithms for space targets.
- 14 September 2021
- Published by Springer Nature in Neural Computing and Applications
- Vol. 34 (2) , 1623-1634
Multi-features guidance network for partial-to-partial point cloud registration
- 14 September 2021
- Published by Springer Nature in Neural Computing and Applications
- Vol. 34 (2) , 1623-1634
Multi-features guidance network for partial-to-partial point cloud registration
- Open Access
Multi-Features Guidance Network for partial-to-partial point cloud registration
To eliminate the problems of large dimensional differences, big semantic gap, and mutual interference caused by hybrid features, in this paper, we propose a novel Multi-Features Guidance Network for partial-to-partial point cloud registration(MFG). The proposed network mainly includes four parts: keypoints' feature extraction, correspondences searching, correspondences credibility computation, and SVD, among which correspondences searching and correspondence credibility computation are the cores of the network. Unlike the previous work, we utilize the shape features and the spatial coordinates to guide correspondences search independently and fusing the matching results to obtain the final matching matrix. In the correspondences credibility computation module, based on the conflicted relationship between the features matching matrix and the coordinates matching matrix, we score the reliability for each correspondence, which can reduce the impact of mismatched or non-matched points. Experimental results show that our network outperforms the current state-of-the-art while maintaining computational efficiency.
Open AccessMulti-Features Guidance Network for partial-to-partial point cloud registration
To eliminate the problems of large dimensional differences, big semantic gap, and mutual interference caused by hybrid features, in this paper, we propose a novel Multi-Features Guidance Network for partial-to-partial point cloud registration(MFG). The proposed network mainly includes four parts: keypoints' feature extraction, correspondences searching, correspondences credibility computation, and SVD, among which correspondences searching and correspondence credibility computation are the cores of the network. Unlike the previous work, we utilize the shape features and the spatial coordinates to guide correspondences search independently and fusing the matching results to obtain the final matching matrix. In the correspondences credibility computation module, based on the conflicted relationship between the features matching matrix and the coordinates matching matrix, we score the reliability for each correspondence, which can reduce the impact of mismatched or non-matched points. Experimental results show that our network outperforms the current state-of-the-art while maintaining computational efficiency.