ImPosing: Implicit Pose Encoding for Efficient Visual Localization
This addresses the challenge of efficient and compact visual localization for autonomous vehicles or robotics in large-scale urban settings, representing a novel method for a known bottleneck.
The paper tackles the problem of real-time visual localization for vehicles in city-scale environments by proposing ImPosing, a method that embeds images and camera poses into a common latent representation to compute similarity scores, achieving significant improvements in accuracy and computational efficiency over prior work.
We propose a novel learning-based formulation for visual localization of vehicles that can operate in real-time in city-scale environments. Visual localization algorithms determine the position and orientation from which an image has been captured, using a set of geo-referenced images or a 3D scene representation. Our new localization paradigm, named Implicit Pose Encoding (ImPosing), embeds images and camera poses into a common latent representation with 2 separate neural networks, such that we can compute a similarity score for each image-pose pair. By evaluating candidates through the latent space in a hierarchical manner, the camera position and orientation are not directly regressed but incrementally refined. Very large environments force competitors to store gigabytes of map data, whereas our method is very compact independently of the reference database size. In this paper, we describe how to effectively optimize our learned modules, how to combine them to achieve real-time localization, and demonstrate results on diverse large scale scenarios that significantly outperform prior work in accuracy and computational efficiency.