Xiao Tan

Papers

Total Citations

H-Index

About

Xiao Tan is a computer vision researcher whose work centers on 3D object detection, with a particular focus on advancing autonomous driving and robotics applications. His most recognized contribution is ZoomNet, a part-aware adaptive zooming neural network designed to tackle one of the field's most persistent challenges: accurately estimating the 3D pose of distant and occluded objects in stereo imagery. By introducing an adaptive zooming mechanism that leverages part-level awareness, ZoomNet demonstrated meaningful improvements in detection robustness under difficult real-world conditions, earning over 80 citations and establishing itself as a noteworthy reference in stereo-based 3D perception. Tan's more recent work, "Coupling and Decoupling: Towards Temporal Feedback for 3D Object Detection" (2025), reflects an evolving research agenda that explores how temporal contextual information from sequential data can be systematically integrated into detection pipelines — a frontier problem as autonomous systems demand richer scene understanding over time. Across his publications, Tan consistently addresses the gap between controlled benchmark performance and the complexities of real-world deployment, making his research directly relevant to practitioners and students working at the intersection of deep learning and embodied intelligence.