LiePoseNet: Heterogeneous Loss Function Based on Lie Group for Significant Speed-up of PoseNet Training Process

TL;DR

This paper introduces a novel Lie group-based heterogeneous loss function for PoseNet that significantly accelerates training from 300 to 10 epochs while maintaining acceptable accuracy in visual localization tasks.

Contribution

The paper proposes a new loss function based on Lie groups and Gaussian distribution to improve training speed and accuracy in PoseNet for visual localization.

Findings

Training speed increased from 300 to 10 epochs

Achieved acceptable localization accuracy

Enhanced geometric awareness in the loss function

Abstract

Visual localization is an essential modern technology for robotics and computer vision. Popular approaches for solving this task are image-based methods. Nowadays, these methods have low accuracy and a long training time. The reasons are the lack of rigid-body and projective geometry awareness, landmark symmetry, and homogeneous error assumption. We propose a heterogeneous loss function based on concentrated Gaussian distribution with the Lie group to overcome these difficulties. Following our experiment, the proposed method allows us to speed up the training process significantly (from 300 to 10 epochs) with acceptable error values.