Learning cross-domain neural networks for sketch-based 3D shape retrieval

Fan Zhu, Jin Xie, Yi Fang

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

Abstract

Sketch-based 3D shape retrieval, which returns a set of relevant 3D shapes based on users' input sketch queries, has been receiving increasing attentions in both graphics community and vision community. In this work, we address the sketch-based 3D shape retrieval problem with a novel Cross- Domain Neural Networks (CDNN) approach, which is further extended to Pyramid Cross-Domain Neural Networks (PCDNN) by cooperating with a hierarchical structure. In order to alleviate the discrepancies between sketch features and 3D shape features, a neural network pair that forces identical representations at the target layer for instances of the same class is trained for sketches and 3D shapes respectively. By constructing cross-domain neural networks at multiple pyramid levels, a many-To-one relationship is established between a 3D shape feature and sketch features extracted from different scales. We evaluate the effectiveness of both CDNN and PCDNN approach on the extended large-scale SHREC 2014 benchmark and compare with some other well established methods. Experimental results suggest that both CDNN and PCDNN can outperform state-of-The-Art performance, where PCDNN can further improve CDNN when employing a hierarchical structure.

Original languageEnglish (US)
Title of host publication30th AAAI Conference on Artificial Intelligence, AAAI 2016
PublisherAAAI press
Pages3683-3689
Number of pages7
ISBN (Electronic)9781577357605
StatePublished - Jan 1 2016
Event30th AAAI Conference on Artificial Intelligence, AAAI 2016 - Phoenix, United States
Duration: Feb 12 2016Feb 17 2016

Other

Other30th AAAI Conference on Artificial Intelligence, AAAI 2016
CountryUnited States
CityPhoenix
Period2/12/162/17/16

Fingerprint

Neural networks

ASJC Scopus subject areas

  • Artificial Intelligence

Cite this

Zhu, F., Xie, J., & Fang, Y. (2016). Learning cross-domain neural networks for sketch-based 3D shape retrieval. In 30th AAAI Conference on Artificial Intelligence, AAAI 2016 (pp. 3683-3689). AAAI press.

Learning cross-domain neural networks for sketch-based 3D shape retrieval. / Zhu, Fan; Xie, Jin; Fang, Yi.

30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press, 2016. p. 3683-3689.

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

Zhu, F, Xie, J & Fang, Y 2016, Learning cross-domain neural networks for sketch-based 3D shape retrieval. in 30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press, pp. 3683-3689, 30th AAAI Conference on Artificial Intelligence, AAAI 2016, Phoenix, United States, 2/12/16.
Zhu F, Xie J, Fang Y. Learning cross-domain neural networks for sketch-based 3D shape retrieval. In 30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press. 2016. p. 3683-3689
Zhu, Fan ; Xie, Jin ; Fang, Yi. / Learning cross-domain neural networks for sketch-based 3D shape retrieval. 30th AAAI Conference on Artificial Intelligence, AAAI 2016. AAAI press, 2016. pp. 3683-3689
@inproceedings{d75198085f2c4e9fb6925a9f07feeeca,
title = "Learning cross-domain neural networks for sketch-based 3D shape retrieval",
abstract = "Sketch-based 3D shape retrieval, which returns a set of relevant 3D shapes based on users' input sketch queries, has been receiving increasing attentions in both graphics community and vision community. In this work, we address the sketch-based 3D shape retrieval problem with a novel Cross- Domain Neural Networks (CDNN) approach, which is further extended to Pyramid Cross-Domain Neural Networks (PCDNN) by cooperating with a hierarchical structure. In order to alleviate the discrepancies between sketch features and 3D shape features, a neural network pair that forces identical representations at the target layer for instances of the same class is trained for sketches and 3D shapes respectively. By constructing cross-domain neural networks at multiple pyramid levels, a many-To-one relationship is established between a 3D shape feature and sketch features extracted from different scales. We evaluate the effectiveness of both CDNN and PCDNN approach on the extended large-scale SHREC 2014 benchmark and compare with some other well established methods. Experimental results suggest that both CDNN and PCDNN can outperform state-of-The-Art performance, where PCDNN can further improve CDNN when employing a hierarchical structure.",
author = "Fan Zhu and Jin Xie and Yi Fang",
year = "2016",
month = "1",
day = "1",
language = "English (US)",
pages = "3683--3689",
booktitle = "30th AAAI Conference on Artificial Intelligence, AAAI 2016",
publisher = "AAAI press",

}

TY - GEN

T1 - Learning cross-domain neural networks for sketch-based 3D shape retrieval

AU - Zhu, Fan

AU - Xie, Jin

AU - Fang, Yi

PY - 2016/1/1

Y1 - 2016/1/1

N2 - Sketch-based 3D shape retrieval, which returns a set of relevant 3D shapes based on users' input sketch queries, has been receiving increasing attentions in both graphics community and vision community. In this work, we address the sketch-based 3D shape retrieval problem with a novel Cross- Domain Neural Networks (CDNN) approach, which is further extended to Pyramid Cross-Domain Neural Networks (PCDNN) by cooperating with a hierarchical structure. In order to alleviate the discrepancies between sketch features and 3D shape features, a neural network pair that forces identical representations at the target layer for instances of the same class is trained for sketches and 3D shapes respectively. By constructing cross-domain neural networks at multiple pyramid levels, a many-To-one relationship is established between a 3D shape feature and sketch features extracted from different scales. We evaluate the effectiveness of both CDNN and PCDNN approach on the extended large-scale SHREC 2014 benchmark and compare with some other well established methods. Experimental results suggest that both CDNN and PCDNN can outperform state-of-The-Art performance, where PCDNN can further improve CDNN when employing a hierarchical structure.

AB - Sketch-based 3D shape retrieval, which returns a set of relevant 3D shapes based on users' input sketch queries, has been receiving increasing attentions in both graphics community and vision community. In this work, we address the sketch-based 3D shape retrieval problem with a novel Cross- Domain Neural Networks (CDNN) approach, which is further extended to Pyramid Cross-Domain Neural Networks (PCDNN) by cooperating with a hierarchical structure. In order to alleviate the discrepancies between sketch features and 3D shape features, a neural network pair that forces identical representations at the target layer for instances of the same class is trained for sketches and 3D shapes respectively. By constructing cross-domain neural networks at multiple pyramid levels, a many-To-one relationship is established between a 3D shape feature and sketch features extracted from different scales. We evaluate the effectiveness of both CDNN and PCDNN approach on the extended large-scale SHREC 2014 benchmark and compare with some other well established methods. Experimental results suggest that both CDNN and PCDNN can outperform state-of-The-Art performance, where PCDNN can further improve CDNN when employing a hierarchical structure.

UR - http://www.scopus.com/inward/record.url?scp=84990053985&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84990053985&partnerID=8YFLogxK

M3 - Conference contribution

SP - 3683

EP - 3689

BT - 30th AAAI Conference on Artificial Intelligence, AAAI 2016

PB - AAAI press

ER -