论文标题
Wganvo：基于生成对抗网络的单眼视觉探光仪
WGANVO: Monocular Visual Odometry based on Generative Adversarial Networks
论文作者
论文摘要
在这项工作中，我们提出了一种基于深度学习的单眼视觉探光法。特别是，对神经网络进行了训练，可以从图像对中回归姿势估计。使用半监督的方法进行培训。与基于几何形状的单眼方法不同，所提出的方法可以恢复场景的绝对尺度，而没有先验知识也不额外的信息。对系统的评估是在著名的Kitti数据集上进行的，在该数据集中证明它可以实时起作用，并且获得的准确性鼓励继续开发基于深度学习的方法。
In this work we present WGANVO, a Deep Learning based monocular Visual Odometry method. In particular, a neural network is trained to regress a pose estimate from an image pair. The training is performed using a semi-supervised approach. Unlike geometry based monocular methods, the proposed method can recover the absolute scale of the scene without neither prior knowledge nor extra information. The evaluation of the system is carried out on the well-known KITTI dataset where it is shown to work in real time and the accuracy obtained is encouraging to continue the development of Deep Learning based methods.
