Xiaojie Zhang

City University of New York, University of Science and Technology Beijing

Papers

Total Citations

H-Index

About

Xiaojie Zhang’s research bridges the critical gap between advanced video analytics and real-world deployment, with a primary focus on low-latency, edge-based computer vision for disaster response. Her most impactful work, "End-to-End Latency Optimization of Multi-view 3D Reconstruction for Disaster Response" (2022, 10 citations), addresses a pressing challenge: enabling complex 3D scene reconstruction on resource-constrained mobile devices like drones and tablets. By optimizing the entire pipeline—from capture to rendering—she provides a practical framework for first responders using BYOD models, significantly reducing the time needed to generate actionable 3D models of disaster scenes. This contribution is vital for time-critical situational awareness. Earlier in her career, Zhang also made foundational contributions to cellular neural/nonlinear networks (CNNs), establishing two new theorems for robust template design in "Two Theorems on the Robust Designs of a Kind of Uncoupled CNNs with Applications" (2007, 3 citations). This work provided a theoretical backbone for reliable image and signal processing, a precursor to her later applied research. Her trajectory from theoretical CNN robustness to practical, latency-sensitive edge analytics demonstrates a unique ability to translate foundational theory into life-saving technology.