Compressive Self-localization Using Relative Attribute Embedding

TL;DR

This paper explores a novel approach to visual place recognition by using relative attribute-based image embeddings, which are domain-adaptive and orthogonal to traditional absolute attribute embeddings, enhancing localization robustness.

Contribution

It introduces the use of relative attribute embeddings for self-localization, offering a new domain-adaptive and potentially more robust image descriptor method.

Findings

Demonstrates effectiveness of relative attribute embeddings in visual place recognition

Shows improved robustness over traditional absolute attribute methods

Provides a new framework for domain-adaptive image descriptors

Abstract

The use of relative attribute (e.g., beautiful, safe, convenient) -based image embeddings in visual place recognition, as a domain-adaptive compact image descriptor that is orthogonal to the typical approach of absolute attribute (e.g., color, shape, texture) -based image embeddings, is explored in this paper.