Revisit Anything: Visual Place Recognition Via Image Segment Retrieval

Abstract

Accurately recognizing a revisited place is crucial for embodied agents to localize and navigate. This requires visual representations to be distinct, despite strong variations in camera viewpoint and scene appearance. Existing visual place recognition pipelines encode the "whole" image and search for matches. This poses a fundamental challenge in matching two images of the same place captured from different camera viewpoints: "the similarity of what overlaps can be dominated by the dissimilarity of what does not overlap". We address this by encoding and searching for "image segments" instead of the whole images. We propose to use open-set image segmentation to decompose an image into `meaningful' entities (i.e., things and stuff). This enables us to create a novel image representation as a collection of multiple overlapping subgraphs connecting a segment with its neighboring segments, dubbed SuperSegment. Furthermore, to efficiently encode these SuperSegments into compact vector repre

Stats

• citations13
• S2 citations—
• github stars168
• HF likes0
• heat score13.05
• arxiv keygarg2024revisit

Related papers

• Are Local Features All You Need For Cross-domain Visual Place Recognition? (2023)13.80
• Fast, Compact And Highly Scalable Visual Place Recognition Through Sequence-based Matching Of Overloaded Representations (2020)9.41
• Regressing Transformers For Data-efficient Visual Place Recognition (2024)3.58
• Spatio-semantic Convnet-based Visual Place Recognition (2019)10.21
• Graph-based Non-linear Least Squares Optimization For Visual Place Recognition In Changing Environments (2020)7.16
• Eigenplaces: Training Viewpoint Robust Models For Visual Place Recognition (2023)15.46
• Structvpr++: Distill Structural And Semantic Knowledge With Weighting Samples For Visual Place Recognition (2025)3.58
• Focus On Local: Finding Reliable Discriminative Regions For Visual Place Recognition (2025)10.70