Zhiqiang Lou

Papers

Total Citations

H-Index

About

Zhiqiang Lou is an emerging researcher specializing in computer vision and robotics, with a particular focus on depth estimation and stereo matching — foundational challenges in enabling machines to perceive and navigate three-dimensional environments. His most notable contribution, **DEFOM-Stereo: Depth Foundation Model Based Stereo Matching** (2025), represents a significant advance in the field by bridging monocular relative depth estimation with traditional stereo matching pipelines. This work addresses long-standing real-world challenges such as occlusion and non-textured surfaces that have historically limited the accuracy of binocular disparity estimation. By leveraging the remarkable generalization capabilities of monocular depth foundation models, Lou's approach offers a more robust framework for metric depth estimation — a critical capability for autonomous systems and robotic perception. The paper has already garnered over 20 citations across its publications since its 2025 release, a strong indicator of early impact within the community. Lou's research sits at an exciting intersection of deep learning and geometric vision, and his work on integrating foundation models into stereo pipelines positions him as a promising contributor to the next generation of spatial understanding systems.