Shuangzhi Li

Shuangzhi Li

Auto-generated profile, based on publicly available information

Shuangzhi Li

University of Alberta, Edmonton, AB, Canada

Research Focus & Metrics

Academic Career Length

7Years (2020-Present)

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

Representative Publications

View All Publications
16 October 2023 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
SLShuangzhi Li
Shuangzhi Li
ZWZhijie Wang
Zhijie Wang
FJFelix Juefei-Xu
Felix Juefei-Xu
Qing GuoQing Guo
Qing Guo
XLXingyu Li
Xingyu Li
LMLei Ma
Lei Ma
Object detection through LiDAR-based point cloud has recently been important in autonomous driving. Although achieving high accuracy on public benchmarks, the state-of-the-art detectors may still go wrong and cause a heavy loss due to the widespread corruptions in the real world like rain, snow, sensor noise, etc . Nevertheless, there is a lack of a large-scale dataset covering diverse scenes and realistic corruption types with different severities to develop practical and robust point cloud detectors, which is challenging due to the heavy collection costs. To alleviate the challenge and start the first step for robust point cloud detection, we propose the physical-aware simulation methods to generate degraded point clouds under different real-world common corruptions. Then, for the first attempt, we construct a benchmark based on the physical-aware common corruptions for point cloud detectors, which contains a total of 1,122,150 examples covering 7,481 scenes, 25 common corruption types, and 6 severities. With such a novel benchmark, we conduct extensive empirical studies on 12 state-of-the-art detectors that contain 6 different detection frameworks. Thus we get several insight observations revealing the vulnerabilities of the detectors and indicating the enhancement directions. Moreover, we further study the effectiveness of existing robustness enhancement methods based on data augmentation, data denoising, test-time adaptation. The benchmark can potentially be a new platform for evaluating point cloud detectors, opening a door for developing novel robustness enhancement methods.

30 September 2020 in IEEE Access
Verified
Journal Article
Open access
SLShuangzhi Li
Shuangzhi Li
ZDZhongliang Deng
Zhongliang Deng
YLYanxu Liu
Yanxu Liu
EHEnwen Hu
Enwen Hu
Under the navigational process of simultaneous beacon-calibrating and target-positioning, the real-time positioning and calibrating accuracy in indoor scenarios is significantly limited. To solve it, a novel Simultaneous Calibration and Localization (SCAL) algorithm framework for indoor scenarios is proposed. The proposed framework is mainly divided into the Target Localization and Beacon Calibration (TLBC) and the Global Optimization (GO) section: in the TLBC section, under the typical distributed beacon layout, a processing mechanism for synchronously locating the target and calibrating beacons is proposed; in GO section, parallel to TLBC section, a global optimization method based on Geometric Dilution of Precision based on Error-propagation (EGDOP) model is further proposed, and it can utilize the geometric trajectory of target and positioning error of target and beacons to efficiently and simultaneously optimize the positioning of target trajectory and beacons from TLBC section. The numerical simulation results show that the improvement of the GO section on the positioning performance is verified. Compared with the SCAL algorithm based on backward processing with inverse trajectory, the proposed framework has improved the accuracy of target and beacon positioning by 39.06% and 58.01%, respectively, and accuracy of positioning reaches 0.2557m and 0.2437m, respectively, in an actual indoor positioning scene.