Yueh-Cheng Liu

Papers

Total Citations

H-Index

About

Yueh-Cheng Liu is a researcher advancing the frontiers of computer vision and depth estimation, with a focus on bridging the gap between sparse sensor data and dense, high-fidelity 3D perception. His key research areas include guided depth estimation, learnable signal processing, and multi-modal sensor fusion for applications in robotics, augmented reality, and 3D reconstruction. Liu’s most notable contribution is the development of **S³ (Learnable Sparse Signal Superdensity)**, a novel framework introduced in his highly cited 2021 paper (21 citations). This work addresses a critical limitation in dense depth estimation: the low density and spatial imbalance of sparse signals from LiDAR or Radar. By learning to generate a superdense representation from sparse inputs, S³ significantly improves depth accuracy without requiring additional hardware, offering a practical solution for real-world systems. His research demonstrates how intelligent, learnable augmentation of sparse data can unlock richer scene understanding, making depth estimation more robust and efficient. Liu’s work is particularly impactful for autonomous navigation and interactive environments, where reliable depth from limited sensors is essential. As a rising voice in computer vision, his contributions are shaping the next generation of perception systems that seamlessly integrate sparse and dense modalities.