PoseINN: Realtime Visual-based Pose Regression and Localization with Invertible Neural Networks

arXiv:2404.13288v1 Announce Type: cross
Abstract: 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 …

abstract applications arxiv augmented reality cameras computational costs cs.cv cs.ro localization mobile networks neural networks paper reality realtime regression robotics solve sota type visual