Jimin Song
Papers
2
Total Citations
15
H-Index
2
About
Jimin Song is a researcher advancing the frontiers of autonomous systems through innovative work in depth estimation and visual-inertial odometry. Their key research areas include monocular depth perception, sensor fusion, and uncertainty-aware learning for mobile robotics. In a notable 2023 contribution, Song developed a monocular depth estimation method for fisheye cameras using knowledge distillation—a technique that compresses complex models into efficient ones while preserving accuracy. This work, which has garnered 9 citations, addresses a critical need in autonomous driving and robotics for reliable, real-time environmental perception from wide-angle imagery. Building on this, Song’s 2024 paper introduced an uncertainty-aware depth network for visual-inertial odometry, achieving 6 citations by enhancing the robustness of simultaneous localization and mapping (SLAM) through multi-sensor integration with IMUs. This approach directly tackles the challenge of collision avoidance in dynamic environments. Song’s research demonstrates a clear trajectory from efficient depth prediction to resilient localization, marking them as a rising contributor to the practical deployment of autonomous navigation technologies.
Research Focus
Key Achievements
Top Papers
- 1
- 2