Guangpu Zhu
Papers
1
Total Citations
9
H-Index
1
About
Guangpu Zhu is a researcher advancing the field of computer vision, with a primary focus on gaze estimation and human-computer interaction. His most cited work, "Style transformed synthetic images for real world gaze estimation by using residual neural network with embedded personal identities" (2022), tackles a critical challenge in gaze tracking: bridging the gap between synthetic training data and real-world performance. Zhu’s key contribution lies in developing a novel framework that integrates style transfer and residual neural networks, embedding personal identity features to improve the robustness and accuracy of gaze estimation across diverse individuals and environments. This approach addresses the scarcity of labeled real-world gaze data, enabling more practical and scalable eye-tracking systems. With 9 citations, his paper has already drawn attention from researchers seeking to enhance gaze-based interfaces, from assistive technologies to virtual reality. Zhu’s work exemplifies a creative synthesis of deep learning and domain adaptation, offering a pathway toward more reliable, personalized gaze estimation. His research holds promise for applications in accessibility, driver monitoring, and interactive systems, positioning him as an emerging voice in the intersection of synthetic data generation and human-centered AI.
Research Focus
Key Achievements
Top Papers
- 1