Learning Multi-view Camera Relocalization with Graph Neural Networks

Fei Xue, Hongbin Zha

Abstract:

We propose to construct a view graph to excavate the information of the whole given sequence for absolute camera
pose estimation. Specifically, we harness GNNs to model
the graph, allowing even non-consecutive frames to exchange information with each other. Rather than adopting
the regular GNNs directly, we redefine the nodes, edges,
and embedded functions to fit the relocalization task. Redesigned GNNs collaborate with CNNs in guiding knowledge propagation and feature extraction respectively to process multi-view high-dimensional image features iteratively
at different levels. Besides, a general graph-based loss
function beyond constraints between consecutive views is
employed for training the network in an end-to-end fashion. Extensive experiments conducted on both indoor and
outdoor datasets demonstrate that our method outperforms
previous approaches especially in large-scale and challenging scenarios.