Zeqi Gu

I am a fourth year PhD student at Cornell Tech, advised by Prof. Abe Davis and Prof. Noah Snavely. Before that I received my bachlor's degree from Cornell University with double major in computer science and mathematics.

My research interests lie in computer vision and graphics, with a focus on content creation using generative models. My past projects involve 3D scene generation, cartoon animation, style transfer, video decomposition and matting, and adversarial attacks. I've had the fortune to intern at NVIDIA, Adobe, Stanford University, and Uber.

Email  /  CV  /  Github

A 3D scene generation pipeline leveraging the power of LLMs and vision foundation models. Our method works on a diverse range of scene categories and aesthetic styles, and produces modular, editable results by representing each object as a 3D mesh.

A first step towards animating characters with arbitrary topologies by pose-conditioned diffusion models.

Fast, lightweight, and top-performing image-to-image translation method for diffusion models, motivated by signal processing (bilateral filtering).

Extending video matting to scenes with complex foreground-background interactions.

A robot arm using vision and tacile information to roll a dough into a given shape. A course project turned into a paper.

An attack method that fine-tunes an existing adversarial example for greater black-box transferability by increasing its perturbation on a pre-specified layer of the source model.

A framework for measuring dataset granularity. We argue that dataset granularity should depend not only on the data samples and their labels, but also on the distance function we choose.

Thank Jon Barron for sharing his website's source code.