Thomas Enxu Li

Title: Thomas Enxu Li: Point-Cloud Segmentation and Geometry

Introduction

Thomas Enxu Li is an inventor associated with Huawei Technologies Co., Ltd. and based in Mississauga, Canada. His patent work focuses on methods for processing point-cloud data, including semantic and instance segmentation and the calculation of surface normals. The supplied information lists 2 patents.

Latest Patents

- System and Method for Proposal-Free and Cluster-Free Panoptic Segmentation System of Point Clouds. This invention describes projecting a point cloud into a range image, extracting features, and generating a feature map. During downsampling, local geometry is used to scale features. The resulting features support semantic segmentation of the point cloud and segmentation of its individual instances.

- Methods and Systems for Deterministic Calculation of Surface Normal Vectors for Sparse Point Clouds. This invention concerns methods for deterministically calculating surface-normal vectors from sparse point-cloud data.

Career Highlights

Li’s listed patents address challenges in interpreting three-dimensional point-cloud data. One focuses on panoptic segmentation, which combines semantic classification with the identification of individual instances. The other concerns calculating surface normals for sparse point clouds. The supplied information associates his work with Huawei Technologies Co., Ltd.

Collaborations

The listed coworkers are Ryan Razani and Yuan Ren. Their names are provided as collaborators, though specific roles or contributions are not specified.

Conclusion

Thomas Enxu Li’s listed inventions focus on point-cloud analysis, spanning panoptic segmentation and deterministic surface-normal calculation. Together, these patents reflect work on methods for extracting useful structure from three-dimensional data.