Haocheng Wan   
万昊承
I am a Master's student at University College London and received my Bachelor's degree from Hangzhou Dianzi University.
I have wide research interests in 3D vision and deep learning.
Email  / 
Google Scholar  / 
Github
|
|
Publication ( * indicates equal contributions in alphabetical order)
|
|
PatchFormer: An Efficient Point Transformer with Patch Attention
Cheng Zhang*, Haocheng Wan*, Xinyi Shen, Zizhao Wu
IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR), 2022
We propose a new type of attention mechanism and a lightweight block with multiscale features.
[paper][code]
|
|
PVT: Point-Voxel Transformer for 3D Deep Learning
Cheng Zhang*, Haocheng Wan*, Shengqiang Liu, Xinyi Shen, Zizhao Wu
International Journal of Intelligent Systems, 2022
We present a pure-transformer backbone architecture for 3D deep learning with high performance and efficiency.
[paper]
[code]
|
|
Graph-PBN: Graph-based Parallel Branch Network for Efficient Point Cloud Learning
Cheng Zhang, Hao Chen, Haocheng Wan, Ping Yang, Zizhao Wu
Graphical Models, Volume 119, 2022, 101120
We introduce a novel parallel branch structure for 3D deep learning. A new convolution operator were designed for local graph-based networks
[paper]
[code]
|
[Web Cite] |