Xueshuang Xiang
Papers
1
Total Citations
23
H-Index
1
About
Xueshuang Xiang is a leading researcher in computational imaging, with a primary focus on non-line-of-sight (NLOS) recognition and reconstruction. Her work addresses the critical challenge of seeing around corners—a capability with transformative potential for autonomous vehicles, robotic vision, and biomedical imaging. Xiang’s most-cited paper, “Accurate but fragile passive non-line-of-sight recognition” (2021, 23 citations), provides a foundational analysis of the trade-offs between accuracy and robustness in passive NLOS systems. She demonstrates that while deep learning models can achieve high classification accuracy for hidden objects, they remain vulnerable to subtle perturbations in the observed light patterns. This insight has shaped subsequent research into more resilient recognition algorithms. Beyond this work, Xiang has contributed to the broader field of computational photography, exploring how physical priors and neural networks can be combined to infer hidden scenes from indirect light. Her research is notable for bridging theoretical understanding with practical system design, making her a key voice in the rapidly evolving domain of NLOS imaging.
Research Focus
Key Achievements
Top Papers
- 1Accurate but fragile passive non-line-of-sight recognition23 citations · 2021