RGB-D to CAD Retrieval with ObjectNN Dataset

Files

Date

2017

Publisher

The Eurographics Association

Abstract

The goal of this track is to study and evaluate the performance of 3D object retrieval algorithms using RGB-D data. This is inspired from the practical need to pair an object acquired from a consumer-grade depth camera to CAD models available in public datasets on the Internet. To support the study, we propose ObjectNN, a new dataset with well segmented and annotated RGB-D objects from SceneNN [HPN 16] and CAD models from ShapeNet [CFG 15]. The evaluation results show that the RGB-D to CAD retrieval problem, while being challenging to solve due to partial and noisy 3D reconstruction, can be addressed to a good extent using deep learning techniques, particularly, convolutional neural networks trained by multi-view and 3D geometry. The best method in this track scores 82% in accuracy.

        @inproceedings{10.2312:3dor.20171048
,
booktitle = {Eurographics Workshop on 3D Object Retrieval
},
editor = {Ioannis Pratikakis and Florent Dupont and Maks Ovsjanikov
},
title = {{RGB-D to CAD Retrieval with ObjectNN Dataset
}},
author = {Hua, Binh-Son and 
Truong, Quang-Trung and 
Johan, Henry and 
Tashiro, Shoki and 
Aono, Masaki and 
Tran, Minh-Triet and 
Pham, Viet-Khoi and 
Nguyen, Hai-Dang and 
Nguyen, Vinh-Tiep and 
Tran, Quang-Thang and 
Phan, Thuyen V. and 
Truong, Bao and 
Tran, Minh-Khoi and 
Do, Minh N. and 
Duong, Anh-Duc and 
Yu, Lap-Fai and 
Nguyen, Duc Thanh and 
Yeung, Sai-Kit and 
Pham, Quang-Hieu and 
Kanezaki, Asako and 
Lee, Tang and 
Chiang, HungYueh and 
Hsu, Winston and 
Li, Bo and 
Lu, Yijuan
},
year = {2017
},
publisher = {The Eurographics Association
},
ISSN = {1997-0471
},
ISBN = {978-3-03868-030-7
},
DOI = {10.2312/3dor.20171048
}
}