ArchitecturalStyle Recognition is a image classifier of 25 different architectural styles, using deep learning.

Made with fastai library. Source code on src. It can be runned in Google Colab

• Dataset
• Training
• Results
• Demonstration
• Run the app locally
• License
• Links

The dataset contains 10113 images from 25 architectural styles. It is a mixed between images scraped from Google Images and the dataset from the paper "Architectural Style Classification using Multinomial Latent Logistic Regression" (ECCV2014), made by Zhe Xu.

The original dataset was made by Zhe Xu. According to the paper, the best accuracy they could get in 2014 was nearly 70% accuracy. Paper "Architectural Style Classification using Multinomial Latent Logistic Regression" (ECCV2014)

• Extended dataset
• Original dataset

The extended dataset is used to test the model with "real world" images, in this case google images, whicha are not taken in perfect conditions.

To make the original dataset bigger and get better results, data augmentation is used. Which is achived creating new images, by transforming (horizontal flip, zoom, light, rotate, etc.) the images.

In this case we go from 4979 to 9588 images.

The dataset is splitted in two:

• training set (80%)
• test set (20%)

We got a 95% of accuracy on this dataset

The app folder contains the application ready to deploy on Render.

The app can be runned locally.

First of all, you have to check all the Requirements, then run this command in your terminal:

python app/server.py serve

Or in python3:

python3 app/server.py serve

Note: it has been only tested in Linux.

MIT License

• fastai
• Render
• Google Colab

Danci, Marian Dumitru (@dumitrux)