Zhi Li (李志)

I am a PhD student at Department of Computer Vision and Machine Learning in Max Planck Institute for Informatics, Saarbrücken, Germany, supervised by Prof. Dr. Bernt Schiele, where I work on domain adaptation for autonomous driving. Prior to that, I worked as a research scientist at Department of Visual Computing and Artificial Intelligence in Max Planck Institute for Informatics, collaborating with Prof. Dr. Christian Theobalt and Prof. Dr. Bernt Schiele on 3D human motion capture in deformable scenes.

I got my master's degree in School of Software Engineering, Xi'an Jiaotong University with a thesis project on semi-supervised 2D and 3D human pose estimation, and my bachelor's degree in School of Foreign Language Studies, Xi'an Jiaotong University with a thesis project on authorship attribution.

My research interests lie in image/video perception for autonomous driving and 3D computer vision.

Email  /  CV  /  Google Scholar  /  Github

A test-time domain adaptation framework for monocular depth estimation that achieves state-of-the-art performance in long-term, ever-changing environments.

A new approach for 3D human motion capture which is aware of the rigid scene geometry.

The first approach for 3D human motion capture making aware of the deformable scene geometry.

An effective approach for estimating absolute-scale human poses in the challenging multi-person setting.

A semi-supervised approach to improve the performance of 3D human pose estimation on monocular videos.

Template from this amazing website.