UniPR-3D: Towards Universal Visual Place Recognition with Visual Geometry Grounded Transformer

TL;DR

UniPR-3D introduces a novel multi-view visual place recognition architecture that leverages 3D and 2D features through a geometry-grounded transformer, significantly improving generalization and state-of-the-art performance.

Contribution

It is the first VPR method to effectively integrate multi-view 3D representations with dedicated feature aggregation modules for enhanced recognition.

Findings

Sets new state-of-the-art performance on VPR benchmarks.

Outperforms single- and multi-view baselines.

Demonstrates strong generalization across diverse environments.

Abstract

Visual Place Recognition (VPR) has been traditionally formulated as a single-image retrieval task. Using multiple views offers clear advantages, yet this setting remains relatively underexplored and existing methods often struggle to generalize across diverse environments. In this work we introduce UniPR-3D, the first VPR architecture that effectively integrates information from multiple views. UniPR-3D builds on a VGGT backbone capable of encoding multi-view 3D representations, which we adapt by designing feature aggregators and fine-tune for the place recognition task. To construct our descriptor, we jointly leverage the 3D tokens and intermediate 2D tokens produced by VGGT. Based on their distinct characteristics, we design dedicated aggregation modules for 2D and 3D features, allowing our descriptor to capture fine-grained texture cues while also reasoning across viewpoints. To further enhance generalization, we incorporate both single- and multi-frame aggregation schemes, along with a variable-length sequence retrieval strategy. Our experiments show that UniPR-3D sets a new state of the art, outperforming both single- and multi-view baselines and highlighting the effectiveness of geometry-grounded tokens for VPR. Our code and models will be made publicly available on Github https://github.com/dtc111111/UniPR-3D.