Location-specific machine learning models trained on synthetic images represent a scalable paradigm for quantifying biodiversity. Digital images of specimens or samples can be manipulated and combined with computer-generated scene elements to produce nearly infinite synthetic images for training machine learning models. Application of this method for the automatic cataloging of wildlife from camera trap images is of urgent need during this time of high extinction rate and environmental change.

Github Repository 

Dataset : Creating Site Specific Synthetic ML Training Datasets for Conservation: Sample Synthetic Training Dataset