Learning a Dynamic Map of Visual Appearance

TL;DR

This paper introduces a method to create a global, dynamic map of visual appearance using billions of geo-tagged images, enabling detailed understanding of appearance variations over time and space without manual annotation.

Contribution

The paper presents a novel framework that constructs a comprehensive, time-aware visual appearance map from large-scale, unlabeled imagery data, integrating overhead imagery with location and time metadata.

Findings

Supports image-driven mapping and geolocalization

Enables verification of image metadata

Operates without manual data annotation

Abstract

The appearance of the world varies dramatically not only from place to place but also from hour to hour and month to month. Every day billions of images capture this complex relationship, many of which are associated with precise time and location metadata. We propose to use these images to construct a global-scale, dynamic map of visual appearance attributes. Such a map enables fine-grained understanding of the expected appearance at any geographic location and time. Our approach integrates dense overhead imagery with location and time metadata into a general framework capable of mapping a wide variety of visual attributes. A key feature of our approach is that it requires no manual data annotation. We demonstrate how this approach can support various applications, including image-driven mapping, image geolocalization, and metadata verification.