Zhiyong Liu

CROSS redefines cross-view localization, enabling models to learn stable semantics and transferable geometry across extreme viewpoint changes, achieving state-of-the-art results.

Dynamic gating allows DGSeg to filter out noise and ambiguity in segmentation cues, leading to superior performance in reasoning segmentation tasks.

Achieving nearly one-pixel precision in lunar rover localization without GNSS signals could revolutionize autonomous exploration on the Moon.