|
Haoqian Wu
Email: wuhaoqian [AT] corp [DOT] netease [DOT] com or woolsey [AT] zju [DOT] edu [DOT] cn
Google Scholar
 / 
Github
I am a Research Scientist at NetEase Fuxi AI Lab, working on 3D computer vision and content generation. I recieved my
Master's degree from Zhejiang University, advised by Prof. Xi Li.
Before that, I received my undergraduate degree in the CS Department, Zhejiang University in July 2018. My research interest lies at the intersection of computer vision and computer graphics. I am particularly excited about neural rendering, 3D reconstruction, differentiable rendering, montion capture and motion generation, etc.
|
Research
|
[3]  NeFII: Inverse Rendering for Reflectance Decomposition with Near-Field Indirect Illumination
Haoqian Wu,
Zhipeng Hu1,2,
Lincheng Li,
Yongqiang Zhang,
Changjie Fan,
Xin Yu
CVPR, 2023 
paper /
project page
We introduce the Monte Carlo sampling based path tracing and cache the indirect illumination as neural radiance, enabling a physics-faithful and easy-to-optimize inverse rendering method.
|
[2]  Towards Unbiased Volume Rendering of Neural Implicit Surfaces with Geometry Priors
Yongqiang Zhang,
Zhipeng Hu,
Haoqian Wu,
Minda Zhao,
Lincheng Li,
Zhengxia Zou,
Changjie Fan
CVPR, 2023 
project page
We revise and provide an additional condition for the unbiased volume rendering. Following this analysis, we propose a new rendering method by scaling the SDF field with the angle between the viewing direction and the surface normal vector.
|
[1]  Condition-Aware Comparison Scheme for Gait Recognition
Haoqian Wu,
Tian Jian,
Yongjian Fu,
Bin Li,
Xi Li
TIP, 2020 
paper
We propose a condition-aware comparison scheme to measure gait pairs’ similarity via a novel module named Instructor. Also, we present a geometry-guided data augmentation approach (Dresser) to enrich dressing conditions. Furthermore, to enhance the gait representation, we propose to model temporal local information from coarse to fine.
|
|