Efficient 3D Pose Estimation and 3D Model Retrieval

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Retrieving 3D models for objects in 2D images is an increasingly important problem, driven by the recent emergence of large databases of 3D models such as ShapeNet. However, this task is challenging for two main reasons: (1) 2D images and 3D models have considerably different representations and characteristics, making it hard to compare them. (2) The appearance of objects can significantly vary with the pose, but it is in general unknown and, thus, multiple poses have to be considered, which is very inefficient. To overcome these problems, we propose first to predict the object's pose and then to use the estimated pose as a prior to retrieve 3D models from a database. In the following, we give a short summary of the approach in Sec.~\ref{sec:3D-pose} and a sketch of results in Sec.~\ref{sec:results}. For more details, we refer to Grabner et.al. "3D Pose Estimation and 3D Model Retrieval for Objects in the Wild".
LanguageEnglish
Title of host publicationProceedings of the 42nd OAGM Workshop
EditorsMartin Welk, Martin Urschler, Peter M. Roth
PublisherVerlag der Technischen Universität Graz
Number of pages1
StatusPublished - 2018

Cite this

Grabner, A., Roth, P. M., & Lepetit, V. (2018). Efficient 3D Pose Estimation and 3D Model Retrieval. In M. Welk, M. Urschler, & P. M. Roth (Eds.), Proceedings of the 42nd OAGM Workshop Verlag der Technischen Universität Graz.

Efficient 3D Pose Estimation and 3D Model Retrieval. / Grabner, Alexander; Roth, Peter M.; Lepetit, Vincent.

Proceedings of the 42nd OAGM Workshop. ed. / Martin Welk; Martin Urschler; Peter M. Roth. Verlag der Technischen Universität Graz, 2018.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Grabner, A, Roth, PM & Lepetit, V 2018, Efficient 3D Pose Estimation and 3D Model Retrieval. in M Welk, M Urschler & PM Roth (eds), Proceedings of the 42nd OAGM Workshop. Verlag der Technischen Universität Graz.
Grabner A, Roth PM, Lepetit V. Efficient 3D Pose Estimation and 3D Model Retrieval. In Welk M, Urschler M, Roth PM, editors, Proceedings of the 42nd OAGM Workshop. Verlag der Technischen Universität Graz. 2018.
Grabner, Alexander ; Roth, Peter M. ; Lepetit, Vincent. / Efficient 3D Pose Estimation and 3D Model Retrieval. Proceedings of the 42nd OAGM Workshop. editor / Martin Welk ; Martin Urschler ; Peter M. Roth. Verlag der Technischen Universität Graz, 2018.
@inproceedings{0cee69f64dd8441c8e15f26ba90781cc,
title = "Efficient 3D Pose Estimation and 3D Model Retrieval",
abstract = "Retrieving 3D models for objects in 2D images is an increasingly important problem, driven by the recent emergence of large databases of 3D models such as ShapeNet. However, this task is challenging for two main reasons: (1) 2D images and 3D models have considerably different representations and characteristics, making it hard to compare them. (2) The appearance of objects can significantly vary with the pose, but it is in general unknown and, thus, multiple poses have to be considered, which is very inefficient. To overcome these problems, we propose first to predict the object's pose and then to use the estimated pose as a prior to retrieve 3D models from a database. In the following, we give a short summary of the approach in Sec.~\ref{sec:3D-pose} and a sketch of results in Sec.~\ref{sec:results}. For more details, we refer to Grabner et.al. {"}3D Pose Estimation and 3D Model Retrieval for Objects in the Wild{"}.",
author = "Alexander Grabner and Roth, {Peter M.} and Vincent Lepetit",
year = "2018",
language = "English",
editor = "Martin Welk and Martin Urschler and Roth, {Peter M.}",
booktitle = "Proceedings of the 42nd OAGM Workshop",
publisher = "Verlag der Technischen Universit{\"a}t Graz",

}

TY - GEN

T1 - Efficient 3D Pose Estimation and 3D Model Retrieval

AU - Grabner,Alexander

AU - Roth,Peter M.

AU - Lepetit,Vincent

PY - 2018

Y1 - 2018

N2 - Retrieving 3D models for objects in 2D images is an increasingly important problem, driven by the recent emergence of large databases of 3D models such as ShapeNet. However, this task is challenging for two main reasons: (1) 2D images and 3D models have considerably different representations and characteristics, making it hard to compare them. (2) The appearance of objects can significantly vary with the pose, but it is in general unknown and, thus, multiple poses have to be considered, which is very inefficient. To overcome these problems, we propose first to predict the object's pose and then to use the estimated pose as a prior to retrieve 3D models from a database. In the following, we give a short summary of the approach in Sec.~\ref{sec:3D-pose} and a sketch of results in Sec.~\ref{sec:results}. For more details, we refer to Grabner et.al. "3D Pose Estimation and 3D Model Retrieval for Objects in the Wild".

AB - Retrieving 3D models for objects in 2D images is an increasingly important problem, driven by the recent emergence of large databases of 3D models such as ShapeNet. However, this task is challenging for two main reasons: (1) 2D images and 3D models have considerably different representations and characteristics, making it hard to compare them. (2) The appearance of objects can significantly vary with the pose, but it is in general unknown and, thus, multiple poses have to be considered, which is very inefficient. To overcome these problems, we propose first to predict the object's pose and then to use the estimated pose as a prior to retrieve 3D models from a database. In the following, we give a short summary of the approach in Sec.~\ref{sec:3D-pose} and a sketch of results in Sec.~\ref{sec:results}. For more details, we refer to Grabner et.al. "3D Pose Estimation and 3D Model Retrieval for Objects in the Wild".

M3 - Conference contribution

BT - Proceedings of the 42nd OAGM Workshop

PB - Verlag der Technischen Universität Graz

ER -