Xiaze Zhang

Papers

Total Citations

H-Index

About

Xiaze Zhang is pioneering advances in 3D computer vision and autonomous systems, with research spanning adaptive 3D object detection and robust LiDAR-visual SLAM. His work on "AS-Det" introduces active sampling strategies for 3D object detection in point clouds, addressing a critical limitation of existing point-based detectors that rely on simplistic sampling for local representation learning. This contribution enhances detection accuracy for autonomous driving and robotics applications. In "DeepPointMap2," Zhang tackles the challenge of limited feature representation in SLAM by developing neural descriptors that replace hand-crafted features, significantly improving cross-modal fusion between LiDAR and visual data. While his most-cited papers currently hold 2 and 1 citations respectively—reflecting their recency (2024-2025)—the innovative nature of his approaches positions him as an emerging voice in the field. Zhang's work directly addresses real-world deployment challenges, offering more adaptive and robust solutions for autonomous navigation systems. His focus on learning-based methods over traditional heuristics marks a meaningful step forward in making 3D perception systems more reliable for safety-critical applications.