HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Thu, 18 Jul 2024 07:22:23 GMT
access-control-allow-origin: *
strict-transport-security: max-age=31556952
etag: W/"6698c2af-1c89"
expires: Mon, 29 Dec 2025 22:32:58 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: 4FEA:3A7A40:9546DB:A795C0:6952FF42
accept-ranges: bytes
age: 0
date: Mon, 29 Dec 2025 22:22:59 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210076-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1767046979.862456,VS0,VE214
vary: Accept-Encoding
x-fastly-request-id: ad6663023b1a02ebdb3f8d8662bfce7f26f0d802
content-length: 1942
Haoran Liu
Delving into Discrete Normalizing Flows on SO(3) Manifold for Probabilistic
Rotation Modeling
Yulin Liu* ,
Haoran Liu* ,
Yingda Yin* ,
Yang Wang ,
Baoquan Chen † ,
He Wang †
CVPR , 2023
project page
/
code
/
arXiv
Mobius flow is a discrete normalizing flow that can be used to model complex distributions over the
SO(3) manifold.
UniDexGrasp: Universal Robotic Dexterous Grasping via Learning Diverse
Proposal Generation and Goal-Conditioned Policy
Yinzhen Xu* ,
Weikang Wan* ,
Jialiang Zhang* ,
Haoran Liu* ,
Zikang Shan ,
Hao Shen ,
Ruicheng Wang ,
Haoran Geng ,
Yijia Weng ,
Jiayi Chen ,
Tengyu Liu ,
Li Yi ,
He Wang †
CVPR , 2023
project page
/
code
/
arXiv
UniDexGrasp is a two stage pipeline for dexterous grasping, where the first part generates a goal
pose with probabilistic models and the second part grasp the object with reinforcement learning.
