Computing the Stereo Matching Cost with a Convolutional Neural Network. Žbontar, J. & LeCun, Y. 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), June, 2015. arXiv: 1409.4326![Computing the Stereo Matching Cost with a Convolutional Neural Network [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex We present a method for extracting depth information from a rectified image pair. We train a convolutional neural network to predict how well two image patches match and use it to compute the stereo matching cost. The cost is refined by cross-based cost aggregation and semiglobal matching, followed by a left-right consistency check to eliminate errors in the occluded regions. Our stereo method achieves an error rate of 2.61 % on the KITTI stereo dataset and is currently (August 2014) the top performing method on this dataset.
@article{zbontar_computing_2015,
title = {Computing the {Stereo} {Matching} {Cost} with a {Convolutional} {Neural} {Network}},
url = {http://arxiv.org/abs/1409.4326},
doi = {10.1109/CVPR.2015.7298767},
abstract = {We present a method for extracting depth information from a rectified image pair. We train a convolutional neural network to predict how well two image patches match and use it to compute the stereo matching cost. The cost is refined by cross-based cost aggregation and semiglobal matching, followed by a left-right consistency check to eliminate errors in the occluded regions. Our stereo method achieves an error rate of 2.61 \% on the KITTI stereo dataset and is currently (August 2014) the top performing method on this dataset.},
urldate = {2022-03-02},
journal = {2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
author = {Žbontar, Jure and LeCun, Yann},
month = jun,
year = {2015},
note = {arXiv: 1409.4326},
keywords = {Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning, Computer Science - Neural and Evolutionary Computing},
pages = {1592--1599},
}
Downloads: 0
{"_id":"GpHs7jen3Yg8xKrAh","bibbaseid":"bontar-lecun-computingthestereomatchingcostwithaconvolutionalneuralnetwork-2015","author_short":["Žbontar, J.","LeCun, Y."],"bibdata":{"bibtype":"article","type":"article","title":"Computing the Stereo Matching Cost with a Convolutional Neural Network","url":"http://arxiv.org/abs/1409.4326","doi":"10.1109/CVPR.2015.7298767","abstract":"We present a method for extracting depth information from a rectified image pair. We train a convolutional neural network to predict how well two image patches match and use it to compute the stereo matching cost. The cost is refined by cross-based cost aggregation and semiglobal matching, followed by a left-right consistency check to eliminate errors in the occluded regions. Our stereo method achieves an error rate of 2.61 % on the KITTI stereo dataset and is currently (August 2014) the top performing method on this dataset.","urldate":"2022-03-02","journal":"2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)","author":[{"propositions":[],"lastnames":["Žbontar"],"firstnames":["Jure"],"suffixes":[]},{"propositions":[],"lastnames":["LeCun"],"firstnames":["Yann"],"suffixes":[]}],"month":"June","year":"2015","note":"arXiv: 1409.4326","keywords":"Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning, Computer Science - Neural and Evolutionary Computing","pages":"1592–1599","bibtex":"@article{zbontar_computing_2015,\n\ttitle = {Computing the {Stereo} {Matching} {Cost} with a {Convolutional} {Neural} {Network}},\n\turl = {http://arxiv.org/abs/1409.4326},\n\tdoi = {10.1109/CVPR.2015.7298767},\n\tabstract = {We present a method for extracting depth information from a rectified image pair. We train a convolutional neural network to predict how well two image patches match and use it to compute the stereo matching cost. The cost is refined by cross-based cost aggregation and semiglobal matching, followed by a left-right consistency check to eliminate errors in the occluded regions. Our stereo method achieves an error rate of 2.61 \\% on the KITTI stereo dataset and is currently (August 2014) the top performing method on this dataset.},\n\turldate = {2022-03-02},\n\tjournal = {2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},\n\tauthor = {Žbontar, Jure and LeCun, Yann},\n\tmonth = jun,\n\tyear = {2015},\n\tnote = {arXiv: 1409.4326},\n\tkeywords = {Computer Science - Computer Vision and Pattern Recognition, Computer Science - Machine Learning, Computer Science - Neural and Evolutionary Computing},\n\tpages = {1592--1599},\n}\n\n","author_short":["Žbontar, J.","LeCun, Y."],"key":"zbontar_computing_2015","id":"zbontar_computing_2015","bibbaseid":"bontar-lecun-computingthestereomatchingcostwithaconvolutionalneuralnetwork-2015","role":"author","urls":{"Paper":"http://arxiv.org/abs/1409.4326"},"keyword":["Computer Science - Computer Vision and Pattern Recognition","Computer Science - Machine Learning","Computer Science - Neural and Evolutionary Computing"],"metadata":{"authorlinks":{}},"html":""},"bibtype":"article","biburl":"https://bibbase.org/zotero/mxmplx","dataSources":["aXmRAq63YsH7a3ufx"],"keywords":["computer science - computer vision and pattern recognition","computer science - machine learning","computer science - neural and evolutionary computing"],"search_terms":["computing","stereo","matching","cost","convolutional","neural","network","žbontar","lecun"],"title":"Computing the Stereo Matching Cost with a Convolutional Neural Network","year":2015}