Estimating ego-pose from cameras is an important problem in robotics with applications ranging from mobile robotics to augmented reality. While SOTA models are becoming increasingly accurate, they can still be unwieldy due to high computational costs. In this paper, we propose to solve the problem by using invertible neural networks (INN) to find the mapping between the latent space of images and poses for a given scene. Our model achieves similar performance to the SOTA while being faster to train and only requiring offline rendering of low-resolution synthetic data. By using normalizing flows, the proposed method also provides uncertainty estimation for the output. We also demonstrated the efficiency of this method by deploying the model on a mobile robot.

翻译：从摄像头估计自我位姿是机器人学中的一个重要问题，其应用涵盖移动机器人到增强现实等领域。尽管当前最先进模型精度不断提升，但由于计算成本高昂，其应用仍可能受限。本文提出利用可逆神经网络（INN）求解该问题，以寻找给定场景中图像隐空间与位姿之间的映射。所提模型在训练速度更快、仅需离线渲染低分辨率合成数据的条件下，仍能达到与当前最先进模型相近的性能。通过使用归一化流，所提方法还能为输出提供不确定性估计。我们通过在移动机器人上部署该模型，进一步验证了该方法的效率。