Mutualvpr: A Mutual Learning Framework For Resolving Supervision Inconsistencies Via Adaptive Clustering

Abstract

Visual Place Recognition (VPR) enables robust localization through image retrieval based on learned descriptors. However, drastic appearance variations of images at the same place caused by viewpoint changes can lead to inconsistent supervision signals, thereby degrading descriptor learning. Existing methods either rely on manually defined cropping rules or labeled data for view differentiation, but they suffer from two major limitations: (1) reliance on labels or handcrafted rules restricts generalization capability; (2) even within the same view direction, occlusions can introduce feature ambiguity. To address these issues, we propose MutualVPR, a mutual learning framework that integrates unsupervised view self-classification and descriptor learning. We first group images by geographic coordinates, then iteratively refine the clusters using K-means to dynamically assign place categories without orientation labels. Specifically, we adopt a DINOv2-based encoder to initial

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keygu2024mutualvpr

Related papers

• Data-efficient Large Scale Place Recognition With Graded Similarity Supervision (2023)16.32
• Embodiedplace: Learning Mixture-of-features With Embodied Constraints For Visual Place Recognition (2025)0.00
• Evaluation Of Visual Place Recognition Methods For Image Pair Retrieval In 3D Vision And Robotics (2026)0.00
• Tightly Coupled Learning Strategy For Weakly Supervised Hierarchical Place Recognition (2022)7.81
• Collaborative Visual Place Recognition Through Federated Learning (2024)2.26
• Mixvpr: Feature Mixing For Visual Place Recognition (2023)22.68
• Focus On Local: Finding Reliable Discriminative Regions For Visual Place Recognition (2025)10.70
• Multires-netvlad: Augmenting Place Recognition Training With Low-resolution Imagery (2022)16.01