Relative Pose Regression with Pose Auto-Encoders: Enhancing Accuracy and Data Efficiency for Retail Applications

TL;DR

This paper introduces a novel PAE-based relative pose regression method that refines absolute camera localization, significantly improving accuracy and data efficiency for retail applications without extra storage requirements.

Contribution

It extends Pose Auto-Encoders to relative pose regression and proposes a refinement scheme that enhances localization accuracy with less training data.

Findings

PAE-based RPR outperforms image-based RPR models of similar architecture.

Refinement with PAE-based RPR improves indoor localization accuracy.

Achieves competitive results using only 30% of training data.

Abstract

Accurate camera localization is crucial for modern retail environments, enabling enhanced customer experiences, streamlined inventory management, and autonomous operations. While Absolute Pose Regression (APR) from a single image offers a promising solution, approaches that incorporate visual and spatial scene priors tend to achieve higher accuracy. Camera Pose Auto-Encoders (PAEs) have recently been introduced to embed such priors into APR. In this work, we extend PAEs to the task of Relative Pose Regression (RPR) and propose a novel re-localization scheme that refines APR predictions using PAE-based RPR, without requiring additional storage of images or pose data. We first introduce PAE-based RPR and establish its effectiveness by comparing it with image-based RPR models of equivalent architectures. We then demonstrate that our refinement strategy, driven by a PAE-based RPR, enhances APR localization accuracy on indoor benchmarks. Notably, our method is shown to achieve competitive performance even when trained with only 30% of the data, substantially reducing the data collection burden for retail deployment. Our code and pre-trained models are available at: https://github.com/yolish/camera-pose-auto-encoders