Panwang Pan

I am currently employed as a Researcher and Developer at Pico Architecture Group within ByteDance Ltd. Previously, I held the position of Senior Algorithm Engineer at Alibaba Cloud, where I specialized in 3D Face Reconstruction and 6DoF Pose Estimation.

In 2019, I earned my Master's degree from Xiamen University, where I was enrolled in the School of Informatics.

My research interests primarily lie at the intersection of Efficient Deep Learning (Quantization & Reparameterization) and 3D Computer Vision (Vision and depth sensor-based 3D reconstruction, understanding and manipulation).

Looking for self-motivated research interns for publishing top-tier papers. Feel free to drop me an email if you are interested in the topics above.

Email / Google Scholar / Github / Twitter

Projects

HumanSplat: Generalizable Single-Image Human Gaussian Splatting with Structure Priors

Panwang Pan*, Zhou Su^* Chenguo Lin^*, Zhen Fan, Yongjie Zhang, Zeming Li¹, Tingting Shen, Yadong Mu, Yebin Liu^‡

We present HumanSplat that predicts the 3D Gaussian Splatting properties of any human from a single input image in a generalizable manner. In particular, HumanSplat comprises a 2D multi-view diffusion model and a latent reconstruction transformer with human structure priors that adeptly integrate geometric priors and semantic features within a unified framework.

4K4DGEN: Panoramic 4D Generation at 4K Resolution

Renjie Li, Panwang Pan*, ,Bangbang Yang, Dejia Xu, Shijie Zhou, Xuanyang Zhang, Zeming Li, Achuta Kadambi, Zhangyang Wang, Zhengzhong Tu, Zhiwen Fan

4K4DGEN accepts a static panoramic image at a resolution of 4096x2048. Upon user interaction through designated clicks, it animates specific regions and converts them into 4D point-based representations.

Learning to Estimate 6DoF Pose from Limited Data: A Few-Shot, Generalizable Approach using RGB Images
Panwang Pan*, Zhiwen Fan*, Brandon Y. Feng, Peihao Wang, Chenxin Li, Zhangyang Wang^†

[Paper] [Project] [Code]

By discretizing the pose search range using multiple pose bins and progressively narrowing the pose search range in each stage using predictions from the previous stage, Cas6D can overcome the large gap between pose candidates and ground truth poses, which is a common failure mode in sparse-view scenarios.

POPE: 6-DoF Promptable Pose Estimation of Any Object, in Any Scene, with One Reference
Panwang Pan*, Zhiwen Fan*, Peihao Wang, Yifan Jiang , Dejia Xu, Hanwen Jiang,
Zhangyang Wang^†

[Paper] [Project] [Code]

We introduce Promptable Object Pose Estimator (POPE), a zero-shot framework for estimating the 6DoF pose of objects in any target image, leveraging a single reference such as a cropped image or a sketch.

StegaNeRF: Embedding Invisible Information within Neural Radiance Fields
Chenxin Li*, Brandon Y. Feng*, Zhiwen Fan*, Panwang Pan, Zhangyang Wang^†

[Paper] [Project] [Code]

StegaNeRF achieves reliable recovery of hidden information with minimal impact on the NeRF rendering quality. This work offers a promising outlook on ownership identification in NeRF and calls for more attention and effort on related problems.

Experience

ByteDance Ltd, Beijing, China, Senior Computer Vision Algorithm Engineer, advised by Cheng Chen and Zeming Li.	08/2022 - Present
Alibaba Cloud, Hangzhou, China, Senior Computer Vision Algorithm Engineer	07/2019 - 07/2022
DevTech Compute, NVIDIA, Beijing, China, AI Developer Technology Engineer Intern advised by Xipeng Li .	07/2018 - 10/2018

Awards

2019, Outstanding Graduates of Xiamen University

2018, National Scholarship for Postgraduates

2018, First Prize of GEDC, Second Prize of MCM & CPIPC

2017, ZhongXian Huang Scholarship of Xiamen University

2015, National Scholarship for Undergraduates