TY - JOUR
T1 - Deep Correlated Holistic Metric Learning for Sketch-Based 3D Shape Retrieval
AU - Dai, Guoxian
AU - Xie, Jin
AU - Fang, Yi
N1 - Funding Information:
Manuscript received May 29, 2017; revised December 10, 2017 and January 31, 2018; accepted March 3, 2018. Date of publication March 19, 2018; date of current version April 12, 2018. This work was supported by the ADEC Award for Research Excellence 2015, titled “Deep Cross-Domain Model for Conceptual Design.” The associate editor coordinating the review of this manuscript and approving it for publication was Prof. Jianfei Cai. (Corresponding author: Yi Fang.) The authors are with the NYU Multimedia and Visual Computing Lab and the Department of Electrical and Computer Engineering, New York University Abu Dhabi 129188, United Arab Emirates, and also with the Department of Electrical and Computer Engineering, New York University Tandon School of Engineering, Brooklyn, NY 11201 USA (e-mail: yfang@nyu.edu).
Publisher Copyright:
© 2018 IEEE.
PY - 2018/7
Y1 - 2018/7
N2 - How to effectively retrieve desired 3D models with simple queries is a long-standing problem in computer vision community. The model-based approach is quite straightforward but nontrivial, since people could not always have the desired 3D query model available by side. Recently, large amounts of wide-screen electronic devices are prevail in our daily lives, which makes the sketch-based 3D shape retrieval a promising candidate due to its simpleness and efficiency. The main challenge of sketch-based approach is the huge modality gap between sketch and 3D shape. In this paper, we proposed a novel deep correlated holistic metric learning (DCHML) method to mitigate the discrepancy between sketch and 3D shape domains. The proposed DCHML trains two distinct deep neural networks (one for each domain) jointly, which learns two deep nonlinear transformations to map features from both domains into a new feature space. The proposed loss, including discriminative loss and correlation loss, aims to increase the discrimination of features within each domain as well as the correlation between different domains. In the new feature space, the discriminative loss minimizes the intra-class distance of the deep transformed features and maximizes the inter-class distance of the deep transformed features to a large margin within each domain, while the correlation loss focused on mitigating the distribution discrepancy across different domains. Different from existing deep metric learning methods only with loss at the output layer, our proposed DCHML is trained with loss at both hidden layer and output layer to further improve the performance by encouraging features in the hidden layer also with desired properties. Our proposed method is evaluated on three benchmarks, including 3D Shape Retrieval Contest 2013, 2014, and 2016 benchmarks, and the experimental results demonstrate the superiority of our proposed method over the state-of-the-art methods.
AB - How to effectively retrieve desired 3D models with simple queries is a long-standing problem in computer vision community. The model-based approach is quite straightforward but nontrivial, since people could not always have the desired 3D query model available by side. Recently, large amounts of wide-screen electronic devices are prevail in our daily lives, which makes the sketch-based 3D shape retrieval a promising candidate due to its simpleness and efficiency. The main challenge of sketch-based approach is the huge modality gap between sketch and 3D shape. In this paper, we proposed a novel deep correlated holistic metric learning (DCHML) method to mitigate the discrepancy between sketch and 3D shape domains. The proposed DCHML trains two distinct deep neural networks (one for each domain) jointly, which learns two deep nonlinear transformations to map features from both domains into a new feature space. The proposed loss, including discriminative loss and correlation loss, aims to increase the discrimination of features within each domain as well as the correlation between different domains. In the new feature space, the discriminative loss minimizes the intra-class distance of the deep transformed features and maximizes the inter-class distance of the deep transformed features to a large margin within each domain, while the correlation loss focused on mitigating the distribution discrepancy across different domains. Different from existing deep metric learning methods only with loss at the output layer, our proposed DCHML is trained with loss at both hidden layer and output layer to further improve the performance by encouraging features in the hidden layer also with desired properties. Our proposed method is evaluated on three benchmarks, including 3D Shape Retrieval Contest 2013, 2014, and 2016 benchmarks, and the experimental results demonstrate the superiority of our proposed method over the state-of-the-art methods.
KW - Sketch-based 3D shape retrieval
KW - deep correlated holistic metric learning
KW - discrepancy across different domains
KW - mitigate
UR - http://www.scopus.com/inward/record.url?scp=85044095838&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85044095838&partnerID=8YFLogxK
U2 - 10.1109/TIP.2018.2817042
DO - 10.1109/TIP.2018.2817042
M3 - Article
C2 - 29671741
AN - SCOPUS:85044095838
VL - 27
SP - 3374
EP - 3386
JO - IEEE Transactions on Image Processing
JF - IEEE Transactions on Image Processing
SN - 1057-7149
IS - 7
ER -