Skip to main content
. 2024 Dec 9;11:1442634. doi: 10.3389/fvets.2024.1442634

Figure 3.

Figure 3

AI Pipeline Overview. The input is a cat facial image, which can be annotated either manually or using an automated landmark detector. The intermediate stage is 48 landmark coordinates, which are then fed to a machine learning model for classification. RF, Random Forest; TPOT, Tree-Based Pipeline Optimization Tool; SciMed, Scientist-Machine Equation Detector; NN, neural network; AE, AutoEncoder.