ArchiTective: Architectural Style Image Classification Model Classification Dataset and Pre-Trained Model by Test 1 ArchiTective MVP

Images

1800 images

Explore Dataset

Model Overview

Model Type: Single Label Classification Model Sample Size per category: 200 images Data Split: 70% training, 20% validation, 10% testing

Purpose: Help identify the architectural styles of buildings. This model focuses on urban environments from Western/Eurocentric locations.

Scope: The goal is to be utilized as a regional tool but there is potential for global application. Should the model properly learn the properties of each architectural style it would be expected to be functional on a global scale regardless of regional aesthetics.

Audience: Intended for a general audience. More specifically could be utilized by either students, tourists, or even developers.

Ideas: Utilized in a guided tour, marketed as a way to better understand the history/culture of a city. Students that study abroad can utilize it for assignments. Developers could utilize it when scouting potential sites of interest.

All in all the goal is to help people better understand the urban landscape. The model predicts the architectural style of a building. Currently is only trained on 4 aesthetics: Art Nouveau; Gothic; Deconstructivism; and Palladian. In later iterations the goal is to have a total of 10 categories.

Trained Model API

This project has a trained model available that you can try in your browser and use to get predictions via our Hosted Inference API and other deployment methods.

Cite This Project

If you use this dataset in a research paper, please cite it using the following BibTeX:

@misc{
                            architective-architectural-style-image-classification-model_dataset,
                            title = { ArchiTective: Architectural Style Image Classification Model Dataset },
                            type = { Open Source Dataset },
                            author = { Test 1 ArchiTective MVP },
                            howpublished = { \url{ https://universe.roboflow.com/test-1-architective-mvp/architective-architectural-style-image-classification-model } },
                            url = { https://universe.roboflow.com/test-1-architective-mvp/architective-architectural-style-image-classification-model },
                            journal = { Roboflow Universe },
                            publisher = { Roboflow },
                            year = { 2023 },
                            month = { apr },
                            note = { visited on 2024-05-15 },
                            }