AddressCLIP: Empowering Vision-Language Models for City-wide Image Address Localization

TL;DR

AddressCLIP introduces an end-to-end vision-language framework for city-wide image address localization, leveraging image-text alignment and spatial constraints to improve accuracy over traditional methods.

Contribution

The paper presents AddressCLIP, a novel end-to-end approach for image address localization that combines contrastive learning with spatial manifold constraints, and provides new datasets for this task.

Findings

Outperforms existing transfer learning methods on new datasets

Achieves high accuracy in city-wide address localization

Demonstrates effectiveness through extensive ablations and visualizations

Abstract

In this study, we introduce a new problem raised by social media and photojournalism, named Image Address Localization (IAL), which aims to predict the readable textual address where an image was taken. Existing two-stage approaches involve predicting geographical coordinates and converting them into human-readable addresses, which can lead to ambiguity and be resource-intensive. In contrast, we propose an end-to-end framework named AddressCLIP to solve the problem with more semantics, consisting of two key ingredients: i) image-text alignment to align images with addresses and scene captions by contrastive learning, and ii) image-geography matching to constrain image features with the spatial distance in terms of manifold learning. Additionally, we have built three datasets from Pittsburgh and San Francisco on different scales specifically for the IAL problem. Experiments demonstrate that our approach achieves compelling performance on the proposed datasets and outperforms representative transfer learning methods for vision-language models. Furthermore, extensive ablations and visualizations exhibit the effectiveness of the proposed method. The datasets and source code are available at https://github.com/xsx1001/AddressCLIP.