Wee Sen Lim

Title: Wee Sen Lim and His Work in Motion-Capture-Based Keypoint Detection.

Introduction

Wee Sen Lim is an inventor associated with Nanyang Technological University in Singapore. His patent work addresses the creation of training data for keypoint detection and the prediction of three-dimensional locations for virtual markers on subjects without physical markers.

Latest Patents

Lim is named on 1 patent: “Method and System for Generating a Training Dataset for Keypoint Detection, and Method and System for Predicting 3D Locations of Virtual Markers on a Marker-Less Subject.”

The invention combines optical marker-based motion capture with video cameras that simultaneously record image sequences. Markers placed on a subject’s bone landmarks or keypoints are captured as three-dimensional trajectories. A computer projects those trajectories onto the images to determine marker locations, interpolates three-dimensional positions, and generates a bounding box around the subject. The resulting training dataset includes at least one image, the marker locations, and the bounding box. The patent also describes using a neural network trained on this dataset to predict three-dimensional locations of virtual markers on a marker-less subject.

Career Highlights

Lim’s listed patent connects motion-capture measurements and video imagery to support machine-learning methods for analyzing human movement. His work is associated with Nanyang Technological University.

Collaborations

Lim’s listed coworkers include Prayook Jatesiktat and Wei Tech Ang.

Conclusion

Wee Sen Lim’s patent work brings together motion capture, computer vision, and neural networks to generate training data and estimate three-dimensional keypoint locations without physical markers.