Visual Place Representation and Recognition from Depth Images

TL;DR

This paper introduces a novel place recognition method using depth images by constructing a 3D model and deriving a 2D scene descriptor, demonstrating robustness to appearance and lighting changes.

Contribution

It presents a new scene descriptor based on 3D models from depth videos for improved place recognition.

Findings

The descriptor is effective under varying scene appearances.

The method is robust to lighting variations.

Results show high accuracy in place recognition tasks.

Abstract

This work proposes a new method for place recognition based on the scene architecture. From depth video, we compute the 3D model and we derive and describe geometrically the 2D map from which the scene descriptor is deduced to constitute the core of the proposed algorithm. The obtained results show the efficiency and the robustness of the propounded descriptor to scene appearance changes and light variations.