Zero-shot Building Age Classification from Facade Image Using GPT-4

TL;DR

This work tackles the problem of estimating building age epochs from facade images in the absence of labeled training data by using a zero-shot classifier based on GPT-4 Vision with carefully designed prompts. The FI-London dataset is introduced to evaluate London facade images across 15 age epochs, and a training-free pipeline outputs an age epoch plus a concise rationale. Results show a modest overall accuracy of 39.69% with a mean absolute error of $0.85$ decades, revealing biases toward adjacent epochs and difficulty predicting very old buildings, though some epochs are predicted with higher fidelity. The study demonstrates the potential of training-free vision-language models for geospatial attribute extraction and highlights the need for better prompt design and dataset balancing to improve fine-grained chronology predictions.

Abstract

A building's age of construction is crucial for supporting many geospatial applications. Much current research focuses on estimating building age from facade images using deep learning. However, building an accurate deep learning model requires a considerable amount of labelled training data, and the trained models often have geographical constraints. Recently, large pre-trained vision language models (VLMs) such as GPT-4 Vision, which demonstrate significant generalisation capabilities, have emerged as potential training-free tools for dealing with specific vision tasks, but their applicability and reliability for building information remain unexplored. In this study, a zero-shot building age classifier for facade images is developed using prompts that include logical instructions. Taking London as a test case, we introduce a new dataset, FI-London, comprising facade images and building age epochs. Although the training-free classifier achieved a modest accuracy of 39.69%, the mean absolute error of 0.85 decades indicates that the model can predict building age epochs successfully albeit with a small bias. The ensuing discussion reveals that the classifier struggles to predict the age of very old buildings and is challenged by fine-grained predictions within 2 decades. Overall, the classifier utilising GPT-4 Vision is capable of predicting the rough age epoch of a building from a single facade image without any training.

Figures (8)

• Figure 1: Sample Images contained in FI-London with age epoch
• Figure 2: The prompt for building age classification used in GPT-4 Vision
• Figure 3: Distribution of Building Epochs in FI-London
• Figure 4: Framework for zero-shot classification of age epoch
• Figure 5: Output example of a correct result, a incorrect result and a "hallucination" result