English
 
Help Privacy Policy Disclaimer
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT
  Adjoint Rigid Transform Network: Task-conditioned Alignment of 3D Shapes

Zhou, K., Bhatnagar, B. L., Schiele, B., & Pons-Moll, G. (2021). Adjoint Rigid Transform Network: Task-conditioned Alignment of 3D Shapes. Retrieved from https://arxiv.org/abs/2102.01161.

Item is

Basic

show hide
Genre: Paper
Latex : Adjoint Rigid Transform Network: {T}ask-conditioned Alignment of {3D} Shapes

Files

show Files
hide Files
:
arXiv:2102.01161.pdf (Preprint), 9KB
Name:
arXiv:2102.01161.pdf
Description:
File downloaded from arXiv at 2021-11-22 09:19
OA-Status:
Visibility:
Public
MIME-Type / Checksum:
application/xhtml+xml / [MD5]
Technical Metadata:
Copyright Date:
-
Copyright Info:
-

Locators

show

Creators

show
hide
 Creators:
Zhou, Keyang1, Author           
Bhatnagar, Bharat Lal1, Author           
Schiele, Bernt1, Author           
Pons-Moll, Gerard1, Author           
Affiliations:
1Computer Vision and Machine Learning, MPI for Informatics, Max Planck Society, ou_1116547              

Content

show
hide
Free keywords: Computer Science, Computer Vision and Pattern Recognition, cs.CV
 Abstract: Most learning methods for 3D data (point clouds, meshes) suffer significant
performance drops when the data is not carefully aligned to a canonical
orientation. Aligning real world 3D data collected from different sources is
non-trivial and requires manual intervention. In this paper, we propose the
Adjoint Rigid Transform (ART) Network, a neural module which can be integrated
with a variety of 3D networks to significantly boost their performance. ART
learns to rotate input shapes to a learned canonical orientation, which is
crucial for a lot of tasks such as shape reconstruction, interpolation,
non-rigid registration, and latent disentanglement. ART achieves this with
self-supervision and a rotation equivariance constraint on predicted rotations.
The remarkable result is that with only self-supervision, ART facilitates
learning a unique canonical orientation for both rigid and nonrigid shapes,
which leads to a notable boost in performance of aforementioned tasks. We will
release our code and pre-trained models for further research.

Details

show
hide
Language(s): eng - English
 Dates: 2021-02-012021-03-212021
 Publication Status: Published online
 Pages: 11 p.
 Publishing info: -
 Table of Contents: -
 Rev. Type: -
 Identifiers: arXiv: 2102.01161
URI: https://arxiv.org/abs/2102.01161
BibTex Citekey: Zhou2102.01161
 Degree: -

Event

show

Legal Case

show

Project information

show

Source

show