Papers
1
Total Citations
1
H-Index
1
About
Dr. Mi Luo is a rising researcher in computer vision and embodied AI, whose work focuses on bridging the extreme viewpoint gap between egocentric (first-person) and exocentric (third-person) video data—a critical challenge for augmented reality and robotics. In their highly innovative 2025 paper "Viewpoint Rosetta Stone: Unlocking Unpaired Ego-Exo Videos for View-invariant Representation Learning," Dr. Luo introduced VIEWPOINTROSETTA, a novel framework that leverages large-scale unpaired ego and exo video data to learn clip-level viewpoint-invariant representations. This work, already garnering attention with early citations, addresses a fundamental limitation in how machines understand human actions across different perspectives, enabling more robust perception systems for AR glasses and autonomous agents. By proposing a method to "translate" between these distinct visual domains without requiring costly paired datasets, Dr. Luo's contribution opens new pathways for scalable, view-agnostic learning. Their research sits at the intersection of representation learning, multi-view video understanding, and human activity recognition, promising to advance how AI systems perceive and interact with the world from any vantage point.
Research Focus
Key Achievements
Top Papers
- 1