Zenan(Chris) Li

Nice to meet you, I'm Zenan Li

San Jose, CA, USA

If you have any questions, feel free to contact me by zenanlicareer@gmail.com

My Background / About Me

I’m a Research and Development Engineer at Deepmotion since 2019. I obtained my M.S. Degree supervised by Prof. Mikhail Gofman at California State University, Fullerton in 2019. After that I was supervised by Prof. Libin Liu for 2 years at Deepmotion.

My research interests include Computer Vision, Computer Graphic, Human Pose Estimation, Motion Generation, Multimodal Models, MLLM, Video Understanding, Parametric Video Generation, Object Detection and Tracking, Synthetic Data Augmentation to overcome data shortage

Publications/Patents

Enhancing Narratives with SayMotion's text-to-3D animation and LLMs

Siggraph 2024 -- Real-Time Live!

Kevin He, Annette Lapham, and Zenan Li

[Project] [Paper]

Methods and systems for detecting 3d poses from 2d images and editing 3d poses

US Patent

Kaichuan He, Jakub Stepien, Zenan Li, Grzegorz Kabza, Marcin Hulist, Mikolaj Korcz

[Project] [Paper]

Measuring dynamic body movement

US Patent

Jeffery Broderick, Douglas Van Blaricom, Jerome Lisk, Zenan Li, Sukhad Anand

[Project] [Paper]

Multiple-Person 3D Human Mesh Recovery from Single 2D Picture

Thesis

Zenan Li

[Paper]

My Work / My Projects *Github Pages bandwidth is low, if you click on READ MORE and get an empty page, please wait until it fully loaded

Text-Video-to-Motion

Keywords: Causal Transformer, Mask Transformer, VQVAE, MultiModal Models

MLLM guided Text-Motion data synthesis

Keywords: MLLM, MultiModal Models, Data Analysis, Data Preprocessing, Clustering

Text-to-Motion Improvements Summary

Keywords: RAG, LLM, Cross Attention

Object Detection & Multi-person tracking

Keywords: Segment Anything, SSD, Person Re-ID, Location Prediction, CLIP, FaceNet

Human(Face+Fingers+Body) Pose Estimation

Keywords: Face, Fingers, Body Pose Estimation, Synthetic Data Augmentation, Diffusion model fine-tune, Discriminator, Motion Deblur, Background Segmentation, Animal

Avatar Video Virtual Try-On

An interesting application I developed. Keywords: Avatar Video Generation(Editing)