Zeyuan Chen
Scholar

Zeyuan Chen

Google Scholar ID: dvplAJkAAAAJ
PhD Student, UC San Diego
Computer Vision
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
910
 
H-index
9
 
i10-index
9
 
Publications
13
 
Co-authors
13
list available
Contact
Emailnnice1216@mail.ustc.edu.cnCVOpen ↗GitHubOpen ↗
Publications
1 items
Soft Tail-dropping for Adaptive Visual Tokenization
2026
Cited
0
Resume (English only)
Academic Achievements
  • - ICCV 2025: X-Dancer: Expressive music to human dance video generation, DepR: Depth Guided Single-view Scene Reconstruction with Instance-level Diffusion, YOLO-Count: Differentiable Object Counting for Text-to-Image Generation
  • - CVPR 2025: X-dyna: Expressive dynamic human image animation
  • - ECCV 2024: Dolfin: Diffusion Layout Transformers without Autoencoder
  • - CVPR 2024: Bayesian Diffusion Models for 3D Shape Reconstruction
  • - AAAI 2024: Bliva: A Simple Multimodal LLM for Better Handling of Text-rich Visual Questions
  • - ICCV 2023: Uni-3D: A Universal Model for Panoptic 3D Scene Reconstruction
  • - Neurips 2022: CASA: Category-agnostic Skeletal Animal Reconstruction
  • - CVPR 2022: VideoINR: Learning Video Implicit Neural Representation for Continuous Space-Time Super-Resolution
  • - CVPR 2021 (Oral): PSD: Principled Synthetic-to-Real Dehazing Guided by Physical Priors
  • - IEEE Transactions on Image Processing (TIP): CERL: A Unified Optimization Framework for Light Enhancement with Realistic Noise
Research Experience
  • - Adobe: Jun. 2025 - Nov. 2025, Mentor: Yuanjun Xiong, Kai Zhang
  • - TikTok: Jun. 2024 - Nov. 2024, Mentor: Hongyi Xu, Linjie Luo
  • - TikTok: Jun. 2023 - Nov. 2023, Mentor: Peng Wang, Kejie Li
  • - University of Illinois Urbana-Champaign: Jun. 2021 - Aug. 2022, Advisor: Shenlong Wang
  • - University of California, San Diego: Apr. 2021 - Nov. 2021, Advisor: Xiaolong Wang
  • - University of Texas at Austin, VITA Group: Jun. 2020 - Mar. 2021, Advisor: Zhangyang (Atlas) Wang
Education
  • Ph.D.: University of California, San Diego, advised by Prof. Zhuowen Tu; B.S.: University of Science and Technology of China, major in Data Science and Big Data Technology.
Background
  • Research interests include generative models, particularly in controllable image, video, and 3D generation. Interned at Adobe Firefly, working with Dr. Yuanjun Xiong and Dr. Kai Zhang on building video generation models.
Miscellany
  • Huge fan of Overwatch and a Tracer One Trick player
Co-authors
13 total
Zhuowen Tu
Zhuowen Tu
Professor, Cognitive Science, Computer Science&Engineering, UC San Diego
Dong Liu
Dong Liu
University of Science and Technology of China
Xiang Zhang
Xiang Zhang
University of California, San Diego
Zhangyang (Atlas) Wang
Zhangyang (Atlas) Wang
XTX Markets & University of Texas at Austin
Yifan Xu
Yifan Xu
Coinbase
Xiaolong Wang
Xiaolong Wang
Associate Professor, UC San Diego
Yinbo Chen
Yinbo Chen
PhD student at UC San Diego
Shenlong Wang
Shenlong Wang
University of Illinois at Urbana-Champaign