UrbanVLP: Multi-Granularity Vision-Language Pretraining for Urban Socioeconomic Indicator Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hao, Xixuan, Chen, Wei, Yan, Yibo, Zhong, Siru, Wang, Kun, Wen, Qingsong, Liang, Yuxuan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915114844684288
author Hao, Xixuan
Chen, Wei
Yan, Yibo
Zhong, Siru
Wang, Kun
Wen, Qingsong
Liang, Yuxuan
author_facet Hao, Xixuan
Chen, Wei
Yan, Yibo
Zhong, Siru
Wang, Kun
Wen, Qingsong
Liang, Yuxuan
contents Urban socioeconomic indicator prediction aims to infer various metrics related to sustainable development in diverse urban landscapes using data-driven methods. However, prevalent pretrained models, particularly those reliant on satellite imagery, face dual challenges. Firstly, concentrating solely on macro-level patterns from satellite data may introduce bias, lacking nuanced details at micro levels, such as architectural details at a place. Secondly, the text generated by the precursor work UrbanCLIP, which fully utilizes the extensive knowledge of LLMs, frequently exhibits issues such as hallucination and homogenization, resulting in a lack of reliable quality. In response to these issues, we devise a novel framework entitled UrbanVLP based on Vision-Language Pretraining. Our UrbanVLP seamlessly integrates multi-granularity information from both macro (satellite) and micro (street-view) levels, overcoming the limitations of prior pretrained models. Moreover, it introduces automatic text generation and calibration, providing a robust guarantee for producing high-quality text descriptions of urban imagery. Rigorous experiments conducted across six socioeconomic indicator prediction tasks underscore its superior performance.
format Preprint
id arxiv_https___arxiv_org_abs_2403_16831
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle UrbanVLP: Multi-Granularity Vision-Language Pretraining for Urban Socioeconomic Indicator Prediction
Hao, Xixuan
Chen, Wei
Yan, Yibo
Zhong, Siru
Wang, Kun
Wen, Qingsong
Liang, Yuxuan
Computer Vision and Pattern Recognition
Artificial Intelligence
Urban socioeconomic indicator prediction aims to infer various metrics related to sustainable development in diverse urban landscapes using data-driven methods. However, prevalent pretrained models, particularly those reliant on satellite imagery, face dual challenges. Firstly, concentrating solely on macro-level patterns from satellite data may introduce bias, lacking nuanced details at micro levels, such as architectural details at a place. Secondly, the text generated by the precursor work UrbanCLIP, which fully utilizes the extensive knowledge of LLMs, frequently exhibits issues such as hallucination and homogenization, resulting in a lack of reliable quality. In response to these issues, we devise a novel framework entitled UrbanVLP based on Vision-Language Pretraining. Our UrbanVLP seamlessly integrates multi-granularity information from both macro (satellite) and micro (street-view) levels, overcoming the limitations of prior pretrained models. Moreover, it introduces automatic text generation and calibration, providing a robust guarantee for producing high-quality text descriptions of urban imagery. Rigorous experiments conducted across six socioeconomic indicator prediction tasks underscore its superior performance.
title UrbanVLP: Multi-Granularity Vision-Language Pretraining for Urban Socioeconomic Indicator Prediction
topic Computer Vision and Pattern Recognition
Artificial Intelligence
url https://arxiv.org/abs/2403.16831