Modelling Uncertainty in Deep Learning for Relocaliz基于不确定性建模的DL相机重定位

本博客仅为作者记笔记之用，不对之处，望见谅，欢迎批评指正。
更多相关博客请查阅：http://blog.csdn.net/weixin_39779106；
如需转载，请附上本文链接：http://blog.csdn.net/weixin_39779106/article/details/79554855

原论文链接
项目链接

一、摘要

原文摘要	翻译
We use a Bayesian convolutional neural network to regress the 6-DOF camera pose from a single RGB image. It is trained in an end-to-end manner with no need of additional engineering or graph optimisation.	我们利用基于贝叶斯的卷积神经网络从单幅彩色图片中回归得到相机六自由度姿态。该方法是一种端到端的方法，不需要多余的工程手段以及后端优化。
taking under 6ms to compute. It obtains approximately 2m and 6$^{\circ}$ accuracy for very large scale outdoor scenes and 0.5m and 10$^{\circ}$ accuracy indoors.	每幅图像的处理时间只要6ms。概算发在室内环境中的精度为2m，6$^{\circ}$，室外环境中的精度为0.5m，10$^{\circ}$。
We leverage the uncertainty measure to estimate metric relocalization error and to detect the presence or absence of the scene in the input image. We show that the model’s uncertainty is caused by images being dissimilar to the training dataset in either pose or appearance.	我们利用的不确定性度量来估计定位误差并检测输入图像场景是否存在。我们展示了该模型的不确定性是由于测试图像与训练数据集在姿态和外观上的不一致所导致的。

二、介绍

原文摘要

翻译

PoseNet removes the need for separate mechanisms for appearance based relocalization and metric pose estimation. Furthermore it does not need to store key frames, or establish frame to frame correspondence. It does this by mapping monocular images to a high dimensional space which is linear in pose and robust to annoyance variables. We can then regress the full 6-DOF camera pose from this representation. This allows us to regress camera pose from the image directly, without the need of tracking or landmark matching.

posenet不需要分别为基于外观的定位和姿态估计设置单独的机制/算法（mechanisms）。此外，它不需要存储关键帧，也不需要建立帧间的联系。它通过将单目图像映射到一个高维线性空间来实现这些关系的构建。因此在这个高维空间中姿态是线性，且对噪声变量具有一定的鲁棒性。然后，我们就可以从这个高维空间描述中回归得到完整的六自由度相机姿态。这使我们能够直接从图像中提取相机的姿态，而不需要进行跟踪或特征匹配。

个人理解1. 这段话很好的总结了作者上一篇文章提出的PoseNet相较于传统定位或位姿估计算法的优势，关于PoseNet的详细解读可以参看本人博客：PoseNet - A Convolutional Network for Real-Time 6-DOF Camera Relocalization 端到端的重定位系统 。

原文摘要	翻译
The main contribution of this paper is extending this framework to a Bayesian model which is able to determine the uncertainty of localization. Our Bayesian convolutional neural network requires no additional memory, and can relocalize in under 6ms per frame on a GPU. By leveraging this probabilistic approach, we