What's New

Synthetic Training Data Used to Create AI Retail Audit System

System classifies products and available shelf space using deep learning

Published Monday April 1, 2019


In this example created by our Deep Vision Data division, a deep learning model based on the ResNet101 architecture was trained to classify product types and available shelf area for a retail store merchandising audit system. The model was trained with 20,000 synthetic product images using a 50-50 split of structured and unstructured domain randomized subsets and an 80-20 training/validation data split. Model validation was also completely done with 100% synthetic training data.

Synthetic training data can be utilized for almost any machine learning application, either to augment a physical dataset or completely replace it. By effectively utilizing domain randomization (DR), synthetic data can be made to be indistinguishable from the physical training information. Synthetic training data is inherently less costly, faster to create, perfectly annotated, and isn’t constrained by availability, time or even the physics of the natural world.

Kinetic Vision President and CEO Rick Schweet explained, "This example demonstrates that synthetic data can be as effective as physical information, or even more so, in training deep learning models. We created the synthetic dataset in days; product photography and manual data annotation would have taken weeks and the resulting dataset wouldn't have the feature variability needed to create a robust, scalable AI system. That's the power of synthetic training data - domain randomization and perfect annotation comes at zero cost." Future plans include integration of the display planogram, which specifies the location and number of specific products, to predict stock outs and merchandising compliance. Learn more at synthetictrainingdata.com.