Haizhuang Liu

Title: Haizhuang Liu: Innovator in 3D Object Detection Technology

Introduction

Haizhuang Liu is a prominent inventor based in Beijing, China. He has made significant contributions to the field of object detection through his innovative methods and systems. His work focuses on enhancing the capabilities of three-dimensional (3D) object detection, which is crucial for various applications in technology and research.

Latest Patents

Haizhuang Liu holds a patent for a "Multimodal weakly-supervised three-dimensional (3D) object detection method and system, and device." This patent describes a method that involves capturing multiple two-dimensional (2D) red, green, and blue (RGB) images using a camera. It also includes acquiring ground points through a vehicle LiDAR sensor and generating a 3D frustum based on 2D box labels from the 2D RGB images. The method filters ground points in the 3D frustum, selects a region with the most 3D points, and generates a 3D pseudo-labeling bounding box of an object. Furthermore, it trains a multimodal superpixel dual-branch network using the 3D pseudo-labeling bounding boxes as labels, with the 2D RGB image and the 3D point cloud as inputs. Finally, it inputs a 2D RGB image of the current frame and a 3D point cloud of the current scenario to generate an overall 3D point cloud.

Career Highlights

Haizhuang Liu is affiliated with the University of Science and Technology Beijing, where he continues to advance his research in 3D object detection technologies. His work has garnered attention for its innovative approach to integrating multimodal data for improved object detection accuracy.

Collaborations

Some of his notable coworkers include Yilin Wang and Rongquan Wang, who have collaborated with him on various projects related to 3D object detection and multimodal systems.

Conclusion

Haizhuang Liu's contributions to the field of 3D object detection exemplify the importance of innovation in technology. His patented methods and systems are paving the way for advancements in how machines perceive and interact with their environments.