Learning confidence measures in the wild
Learning confidence measures in the wild
复制标题
在野外学习信心测量
DOI：
10.5244/c.31.133
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
L. D. Stefano
中科院分区：
文献类型：
--
作者：
Fabio Tosi;Matteo Poggi;S. Mattoccia;A. Tonioni;L. D. Stefano
立体声置信度测度在近年来有关立体声的研究中越来越受欢迎，被有效地用于提高立体声精度。虽然大多数度量是通过处理来自成本量的线索获得的，但表现最好的度量通常利用随机森林或cnn来预测匹配可靠性。因此，需要适当数量的标记数据来有效地训练这些置信度度量。由于在实际应用中并不总是可用这样的真值标签，在本文中，我们提出了一种适合于以自我监督的方式训练置信度度量的方法。利用适当选择的常规测量池，我们自动检测非常可靠像素的子集以及来自立体算法输出的错误样本的子集。该策略为基于机器学习技术的训练置信度度量提供了标签，而不需要真值标签。与最先进的技术相比，我们的方法既不局限于图像序列，也不局限于图像内容。在三个具有挑战性的数据集上使用三种立体算法和基于机器学习技术的三种最先进的置信度度量的实验结果证实了我们提出的自监督训练的有效性。
Confidence measures for stereo earned increasing popularity in most recent works concerning stereo, being effectively deployed to improve its accuracy. While most measures are obtained by processing cues from the cost volume, top-performing ones usually leverage on random-forests or CNNs to predict match reliability. Therefore, a proper amount of labeled data is required to effectively train such confidence measures. Being such ground-truth labels not always available in practical applications, in this paper we propose a methodology suited for training confidence measures in a self-supervised manner. Leveraging on a pool of properly selected conventional measures, we automatically detect a subset of very reliable pixels as well as a subset of erroneous samples from the output of a stereo algorithm. This strategy provides labels for training confidence measures based on machine-learning technique without ground-truth labels. Compared to state-of-the-art, our method is neither constrained to image sequences nor to image content. Experimental results on three challenging datasets with three stereo algorithms and three state-of-the-art confidence measures based on machine-learning techniques confirm the effectiveness of our proposal for self-supervised training.
影响因子：
9.2
作者：
Geiger, A.;Lenz, P.;Urtasun, R.
通讯作者：
Urtasun, R.