Jeff Tan

Hello! I am a second-year CS PhD student at Stanford University, fortunate to be advised by Prof. Jiajun Wu in the Stanford Vision and Learning Lab. I am also a student researcher at Google in San Francisco. Previously, I completed my MS in Robotics and undergrad in CS at Carnegie Mellon University, advised by Prof. Deva Ramanan. My research is supported by the NSF Graduate Research Fellowship (2023-).

I am interested in computer vision and its intersection with robotics, generative modeling, and representation learning.

CV  /  Google Scholar  /  Github  /  Email  /  Twitter

We generate minutes-long physically-plausible 3D animations from text, depicting humanoid robots, people, and mechanisms, with better quality than bare GPT-6 agents.

We recover high-quality and simulation-ready scene geometry from Internet videos, for humanoid real-to-sim within complex terrains.

We reconstruct dynamic 4D human-centric scenes (e.g. playing the piano and bicycle repair) from sparse-view video.

From a set of multi-view images, we learn a denoising diffusion model that outputs scene geometry and cameras in global frame.

From a single monocular video, we reconstruct humans in loose clothing and interacting with objects, using a hierarchical deformation field and image-based priors.

We learn real-time feed-forward pose and shape predictors, by distilling knowledge from offline differentiable rendering optimizers.

A Python library for 4D reconstruction of humans, animals, and scenes from monocular videos.

Credits to Jon Barron for this website's template.