Gege Zhang

Papers

Total Citations

H-Index

About

Gege Zhang is a leading researcher in 3D computer vision and human-robot interaction, with a primary focus on advancing semantic segmentation and natural language understanding for autonomous systems. Their most influential work, "AttAN: Attention Adversarial Networks for 3D Point Cloud Semantic Segmentation" (2020, 19 citations), introduces a novel framework that addresses the critical limitation of independent point predictions in existing segmentation methods. By integrating attention mechanisms with adversarial training, Zhang’s approach significantly improves contextual coherence and accuracy in 3D point cloud analysis, directly impacting applications in autonomous driving, augmented reality, and robotic sensing. More recently, Zhang has expanded into embodied AI with "Referring Expression Comprehension in semi-structured human–robot interaction" (2025), pioneering methods that enable robots to interpret complex natural language commands within dynamic, semi-structured environments. This work bridges the gap between vision and language, enhancing intuitive human-robot collaboration. With a growing citation footprint and contributions that push the boundaries of both theoretical and applied AI, Zhang is establishing a reputation for developing robust, context-aware models that bring machines closer to seamless real-world interaction.