Euler angles based loss function for camera relocalization with Deep learning

TL;DR

This paper introduces an Euler angles based loss function for camera relocalization using deep learning, eliminating the need for parameter selection and achieving competitive results on standard datasets.

Contribution

The paper proposes a novel Euler angles based loss function for camera pose regression that simplifies parameter tuning and maintains competitive accuracy.

Findings

Achieves competitive performance on 7 Scenes and King's College datasets.

Eliminates the need for parameter selection in orientation representation.

Demonstrates the effectiveness of Euler angles in deep learning-based camera relocalization.

Abstract

Deep learning has been applied to camera relocalization, in particular, PoseNet and its extended work are the convolutional neural networks which regress the camera pose from a single image. However there are many problems, one of them is expensive parameter selection. In this paper, we directly explore the three Euler angles as the orientation representation in the camera pose regressor. There is no need to select the parameter, which is not tolerant in the previous works. Experimental results on the 7 Scenes datasets and the King's College dataset demonstrate that it has competitive performances.