معرفی
Fengqing Maggie Zhu is an Associate Professor at the Elmore Family School of Electrical and Computer Engineering within Purdue University, West Lafayette campus. Her research spans image processing, video compression, computer vision, and smart health, with notable contributions to learned image compression, 3D reconstruction, and nutrition analysis via computer vision.
Educational background:
- BS in Electrical Engineering, Purdue University (2004)
- MS in Electrical and Computer Engineering, Purdue University (2006)
- PhD in Electrical and Computer Engineering, Purdue University (2011)
Her work focuses on developing machine learning-based compression techniques for 2D/3D images and videos, with applications in food portion estimation, wearable dietary monitoring, and virtual reality facial expression tracking. She explores structured pruning, mixed precision quantization, and continual learning to create efficient, robust systems for edge-cloud collaboration.
The 2025-2024 article collection reveals concentrated efforts in learned image compression (with 8 papers on quantization, pruning, hierarchical VAEs), food-related computer vision (12+ papers on portion estimation, databases, classification), and 3D reconstruction (MetaFood3D dataset, ICP-3DGS algorithm). Emerging themes include privacy-preserving AI for wearable cameras and class-incremental learning frameworks.
Contact: zhu0@purdue.edu
