Hao Wei

Hao Wei

Auto-generated profile, based on publicly available information

Hao Wei

Institute of Automation, State Key Laboratory of Multimodal Artificial Intelligence Systems, Chinese Academy of Sciences, Beijing, China

Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
21 September 2021 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
HWHao Wei
Hao Wei
FTFulin Tang
Fulin Tang
ZXZewen Xu
Zewen Xu
CZChaofan Zhang
Chaofan Zhang
YWYihong Wu
Yihong Wu
Weak texture and motion blur are always challenging problems for visual-inertial odometry (VIO) systems. To improve accuracy of VIO systems in the challenging scenes, we propose a point-line-based VIO system with novel feature hybrids and with novel predicting-matching for long line track. Point-line features with shorter tracks are categorized into “MSCKF” features and with longer tracks into “SLAM” features. Especially, “SLAM” lines are added into the state vector to improve accuracy of the proposed system. Besides, to ensure the reliability and stability of detection and tracking of line features, we also propose a new “Predicting-Matching” line segment tracking method to increase the track lengths of line segments. Experimental results show that the proposed method outperforms the state-of-the-art methods of VINS-Mono [1], PL-VINS [2] and OpenVINS [3]) on both a public dataset and a collected dataset in terms of accuracy. The collected dataset is full of extremely weak textures and motion blurs. On this dataset, the proposed method also obtains better accuracy than ORB-SLAM3 [4].

27 July 2022 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
HWHao Wei
Hao Wei
FTFulin Tang
Fulin Tang
ZXZewen Xu
Zewen Xu
YWYihong Wu
Yihong Wu
Man-made buildings exhibit structural regularity, which can provide strongly geometrical constraints for Visual-Inertial Odometry (VIO) systems. To make full use of the structural information, we propose a new structural regularity aided VIO with novel coordinate alignment and line triangulation under Manhattan world assumption. The proposed VIO system is built upon OpenVINS [1] and partly based on our previous work [2]. The proposed coordinate alignment method makes the Jacobians and reprojection errors become concise but also makes the required computation number for transformations decrease. In addition, a novel structural line triangulation method is provided, in which the global orientation of a structural line is used to refine its 3D position. All the novelties result in a more accurate and fast VIO system. The system is tested on EuRoC MAV dataset and a self-collected dataset. Experimental results demonstrate that the proposed method obtains better accuracy compared with state-of-the-art (SOTA) point-based systems (VINS-Mono [3] and OpenVINS [1]), point-line-based systems (PL-VINS [4] and Wei et al. [2]), and structural line-based system (StructVIO [5]). Notably, the self-collected dataset is recorded on Manhattan world scenes, and is also full of challenging weak texture and motion blur situations. On the dataset, the accuracy of our method is increased by 40.7% compared with the SOTA point-line-based systems.

16 April 2024 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
HWHaolin Wang
Haolin Wang
HWHao Wei
Hao Wei
ZXZewen Xu
Zewen Xu
ZLZeren Lv
Zeren Lv
PZPengju Zhang
Pengju Zhang
YWYihong Wu
Yihong Wu
Planar structures, prevalent in man-made environments, can be observed by a camera for significant periods of time due to their large spatial presence. These structures provide strong planar regularities for Simultaneous Localization and Mapping (SLAM) systems, facilitating long-term navigation. Therefore, we propose a novel point-plane-based stereo SLAM system, fully regularized by plane features within a unified non-linear optimization framework. The core of our method is an accurate and efficient stereo plane extraction algorithm with strict 2D and 3D outlier rejection mechanisms, effectively extracting main planes from robust stereo correspondences and enabling real-time point-plane association. Furthermore, we introduce a novel optimization formulation, incorporating geometric feature (point and plane) and across-feature (point-on-plane) constraints that promote each other through the mutual constraints between associated point and plane features, which fully exploits plane constraints to improve the performance of SLAM system. The proposed plane extraction algorithm is evaluated on the EuRoC MAV dataset, achieving significant improvements in number, accuracy, reliability, and efficiency over the state-of-the-art (SOTA) stereo point-plane-based system (Zhang et al., 2021). The results of an ablation study on two public datasets show that the proposed SLAM system outperforms (Zhang et al., 2021) in both accuracy and robustness, and further demonstrate the mutual enhancement between the two types of constraints.

14 January 2026 in IEEE Transactions on Robotics
Verified
Journal Article
Closed access
BXBo Xu
Bo Xu
ZXZewen Xu
Zewen Xu
YHYijia He
Yijia He
ZOZhanpeng Ouyang
Zhanpeng Ouyang
HWHao Wei
Hao Wei
HLHongdong Li
Hongdong Li
We propose a novel initialization and online spatial-temporal calibration method for visual-inertial odometry (VIO), which decouples rotation and translation estimation to achieve higher accuracy and better robustness. Existing initialization methods suffer from limited accuracy or robustness (e.g., in scenarios with small translational motion) and rarely integrate simultaneous spatial-temporal calibration during initialization, despite its considerable practical value. Our proposed method leverages rotation-translation decoupling constraints to enable simultaneous estimation of gyroscope bias, extrinsic rotation, and camera-IMU time offset-even under pure rotational motion. Moreover, we are the first to conduct observability analysis on rotational constraints in rotation-translation decoupling methods, experimentally identifying the unobservable state-space directions under three degenerate motions within our approach. We also perform extensive experiments to delineate practical parameter solution boundaries for our method, with both efforts substantially enhancing the overall practical applicability of decoupling-based methods. Extensive experiments on simulated and real-world datasets demonstrate that our method outperforms state-of-the-art approaches in accuracy and robustness while maintaining computational efficiency. Furthermore, experiments verify that it significantly improves convergence in VIO systems.

3 July 2024 in IEEE Signal Processing Letters
Verified
Journal Article
Closed access
SXSiyi Xiang
Siyi Xiang
SGShiyi Guo
Shiyi Guo
HWHao Wei
Hao Wei
BLBingxi Liu
Bingxi Liu
DZDabo Zhang
Dabo Zhang
Point cloud registration plays an important role in many applications of computer vision and robotics. Outlier rejection is an essential step in this task. In this paper, we propose an efficient hand-crafted outlier rejection algorithm for point cloud registration. The proposed method includes four modules: seed selection, consensus set construction and sampling, transformation matrix calculation, and hypothesis selection. Particularly, we propose a novel seed selection module based on the addition of elements in the Spatial Consistency (SC) matrix and a new noise suppression strategy in constructing consensus sets. Experimental results on various datasets show that the proposed algorithm presents superior performance over the previous hand-crafted methods.

17 April 2026 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
SGShiyi Guo
Shiyi Guo
TJTong Jia
Tong Jia
BYBi Yang
Bi Yang
YWYihong Wu
Yihong Wu
HWHao Wei
Hao Wei
NANing An
Ning An
Recently, color point cloud registration has begun to receive attention. Unlike geometric-only point clouds, color point clouds incorporate additional color information. Therefore, color point cloud registration can achieve higher accuracy than geometric-only point cloud registration. Despite the success, existing methods are computationally intensive due to the high resource demands of Transformer. In this paper, we propose a Mamba architecture based registration algorithm, CGFMamba-PCR, with color-geometric fusion. Specifically, we propose CGFMamba, a novel color-geometric fusion Mamba network for color point cloud registration, which enhances feature representation through color-geometric guided point ordering and positional encoding. For the input color point cloud pair, they are passed through CGFMamba based feature extraction module to obtain their corresponding features. Then, these features are passed through feature matching and outlier rejection modules to obtain final registration result. Furthermore, an ordering method for the Mamba architecture is proposed that clusters color hue and sorts spatial coordinates. Experiments on Color3DMatch and Color3DLoMatch datasets demonstrate that the proposed algorithm outperforms the state-of-the-art (SOTA) methods. The code of the proposed algorithm will be open-sourced upon acceptance of this paper.