Unifying Local and Global Multimodal Features for Place Recognition in Aliased and Low-Texture Environments

Abstract-- Perceptual aliasing and weak textures pose significant challenges to the task of place recognition, hindering the performance of Simultaneous Localization and Mapping (SLAM) systems. This paper presents a novel model, called UMF (standing for Unifying Local and Global Multimodal Features) that 1) leverages multi-modality by crossattention blocks between vision and LiDAR features, and 2) includes a re-ranking stage that re-orders based on local feature matching the top-k candidates retrieved using a global representation. UMF outperforms significantly previous baselines in those challenging aliased environments. Simultaneous Localization and Mapping (SLAM) has emerged as a central technology in a multitude of industries In this paper we propose a novel multimodal place recognition including autonomous driving [1], [2], automated method that we denote as UMF, standing for Unifying construction [3], and agriculture [4], [5]. Our model leverages and adoption have been accelerated by advancements in deep local and global features from visual and LiDAR data, sensor technologies, including multi-camera setups, RGB-D fusing both modalities via cross-attention mechanisms.