Ling Wei

WORK HISTORY

EDUCATION

ABOUT LING WEI

Chia-Ling is a Data Scientist at DLR Group. Before joining the company, she finished her PhD program at ASU, focusing on Applied Machine Learning to real-world projects.One of her research projects involved the development of synthetic turbine defect image data using advanced techniques such as Variational Autoencoders (VAE), Generative Adversarial Networks (GAN), and diffusion models. This endeavor allowed for the generation of high-quality synthetic data crucial for enhancing turbine inspection processes.In another project, she conducted large-scale road infrastructure spatial analysis using semi-supervised learning methodologies. This involved collecting city-scale LiDAR sensor data, extracting features from extensive datasets, and implementing image processing techniques to facilitate cyclist safety analysis using YOLOv8 algorithms. Additionally, she explored object detection from indoor LiDAR sensor data and performed wall detection on projected 3D datasets.[PUBLICATION]- Wei, C. & Czerniawski, T.(2023). Plan-View Wall Detection for Indoor Point Clouds using Weak Supervision. Proceedings of the International Conference on Computing in Civil Engineering. Furthermore, she has extensive experience in object detection and instance segmentation, particularly in 2D drawings. She developed machine learning pipelines for 3D model reconstruction via communicating with Revit API & Blender API, handled imbalanced datasets effectively, and automated dataset generation processes to improve labeling efficiency significantly. Her expertise extends to utilizing state-of-the-art models such as YOLOv4, Faster R-CNN, and Mask R-CNN for object detection and instance segmentation tasks.[PUBLICATIONS]- Wei, C, Gupta, M, & Czerniawski, T.(2023). Interoperability between Deep Neural Networks and 3D Architectural Modeling Software: Affordances of Detection and Segmentation. Buildings, Application of Computer Technology in Buildings. Wei, C, Gupta, M, & Czerniawski, T.(2022). Automated Wall Detection in 2D CAD Drawings to create Digital 3D Models. Proceedings of the International Symposium on Automation and Robotics in Construction (IAARC).