Deep Learning Architect - Classification for Architectural Design through the Eye of Artificial Intelligence (2018)
Notes on a 2018 paper by Yuji Yoshimura, Bill Cai, Zhoutong Wang, and Carlo Ratti
Using a convolutional neural network to recognise top-level architects' individual styles. Almost 70% or architects can be distinguished with 80% accuracy. 20,000 sample images by 34 past Prisker Prize winners.
Gradient-weighted Class Activation Mapping (Grad-CAM). Enables us to understand the focus of the machine eye for classification
Dimension reduction and clustering - Linear principal component analysis. k-means clustering
34 architects - all past Prisker Prize winners
20,000 total sample images
Goolge Tensorflow. Two GeForce GTX 1070Ti. Training complete in 8 hours.
Tends to confuse Koolhaas, Holl, Perrault with other architects
Correctly distinguishes Kahn, Siza & van de Rohe
“Top 1 accuracy indicates the probability of whether the image can correctly match with the target label. Conversely, the top 5 accuracy suggests the probability of whether the correct image can appear with the target label among five pictures ordered according to their highest probability.”
Almost 70% of architects can be distinguished with more than 80% probability (Top 5 accuracy)
45% distinguished with 90% probability (Top 5 accuracy)
Indoor scenes more distinguishable to machine eye (~5% more accurate)