My research interests mainly focus on developing novel models and algorithms to address practical challenges in deploying CV/ML systems to various real-world application domains.

I am currently interested and engaged in the following topics:

• Visual geometry & 3D vision: visual correspondence, object/camera pose estimation, multi-view geometry, structure-from-motion, optical flow, 3D reconstruction, 3D symmetry, point cloud registration, robotics, embodied AI, and spatial reasoning.
• Representation learning: invariance and equivariance, geometric deep learning, equivariant learning, self-supervised learning, diffusion models for discriminative tasks, foundation models for image/video/3D.
• Visual perception: object detection/segmentation, video summarization, document understanding, task-specific fine-tuning with multi-modal LLMs.

The application domains of interest include, but are not limited to:

• Autonomous vision systems (e.g., visual SLAM, autonomous driving, and industrial robotics)
• AR/VR technologies (e.g., eye tracking, human-centric AI, wearable devices, and neural rendering)

Jongmin Lee is an assistant professor at Chung-Ang University (CAU) and the principal investigator of the Spatial Intelligence Lab. His research focuses on computer vision and machine learning, with particular interests in visual geometry, 3D vision, geometric deep learning, and spatial reasoning with large language models. He leads research efforts that bridge theoretical advances with real-world applications in robotics, AR/VR, and autonomous vision systems.

At the Spatial Intelligence Lab, his group develops novel models and algorithms for spatial intelligence, including multi-view geometry, structure-from-motion, pose estimation, 3D reconstruction, and embodied AI. The lab also explores representation learning approaches such as invariance, equivariance, and self-supervised methods, with applications ranging from industrial robotics and autonomous driving to wearable AR/VR systems and human-centric AI.