Multitasking Correlation Network for Depth Information Reconstruction

Quang Van Nguyen, Duy Cao Hoang, Phuc Nguyen Hong

Abstract


In this paper, we propose a novel multi-tasking network for stereo matching. The proposed network is trained to approximate similarity functions in statistics and linear algebra such as correlation coefficient, distance correlation and cosine similarity. By doing this, the proposed method decreases the amount of time needed to calculate the disparity map by using CNN's ability to calculate multiple pairs of image patches at the same time. We then compare the execution time and overall accuracy between the traditional method using functions and our method. The results show the model's ability to mimic the traditional method's performance while taking considerably less time to perform the task.

Full Text:

PDF

References


V. Q. Dinh, V. D. Nguyen, and J. W. Jeon. Robust matching cost functionfor stereo correspondence using matching by tone mapping and adaptiveorthogonal integral image.IEEE Transactions on Image Processing,24(12):5416–5431, Dec 2015.

V. Q. Dinh, V. D. Nguyen, H. Van Nguyen, and J. W. Jeon. Fuzzyencoding pattern for stereo matching cost.IEEE Transactions on Circuitsand Systems for Video Technology, 26(7):1215–1228, July 2016.

V. Q. Dinh, C. C. Pham, and J. W. Jeon. Matching cost function usingrobust soft rank transformations.IET Image Processing, 10(7):561–569,2016.

V. Q. Dinh, C. C. Pham, and J. W. Jeon. Robust adaptive normalizedcross-correlation for stereo matching cost computation.IEEE Transac-tions on Circuits and Systems for Video Technology, 27(7):1421–1434,July 2017.

Vinh Quang Dinh, Farzeen Munir, Ahmad Muqeem Sheri, and MoonguJeon.Disparity estimation using stereo images with different focallengths.IEEE Transactions on Intelligent Transportation Systems,21(12):5258–5270, 2020.

John Duchi, Elad Hazan, and Yoram Singer.Adaptive subgradientmethods for online learning and stochastic optimization.Journal ofMachine Learning Research, 12:2121–2159, 07 2011.

Guillaume Lample and Franc ̧ois Charton. Deep learning for symbolicmathematics.CoRR, abs/1912.01412, 2019.

Zewen Li, Wenjie Yang, Shouheng Peng, and Fan Liu. A survey ofconvolutional neural networks: Analysis, applications, and prospects.CoRR, abs/2004.02806, 2020.

Donald St. P. Richards. Distance correlation: A new tool for detectingassociation and measuring correlation between data sets, 2017.

Haohan Wang, Bhiksha Raj, and Eric P. Xing. On the origin of deeplearning.CoRR, abs/1702.07800, 2017.

Yaochen Xie, Zhao Xu, Zhengyang Wang, and Shuiwang Ji.Self-supervised learning of graph neural networks: A unified review.arXivpreprint arXiv:2102.10757, 2021.

Wenpeng Yin, Katharina Kann, Mo Yu, and Hinrich Sch ̈utze. Compar-ative study of CNN and RNN for natural language processing.CoRR,abs/1702.01923, 2017.

Vitalii Zhelezniak, Aleksandar Savkov, April Shen, and Nils Y. Ham-merla. Correlation coefficients and semantic textual similarity.CoRR,abs/1905.07790, 2019.




DOI: http://dx.doi.org/10.21553/rev-jec.293

Copyright (c) 2023 REV Journal on Electronics and Communications


ISSN: 1859-378X

Copyright © 2011-2024
Radio and Electronics Association of Vietnam
All rights reserved