Rendering synthetic imagery from gaming engine environments allows us to create data featuring any number ofobject orientations, conditions, and lighting variations. This capability is particularly useful in classi cation tasks,where there is an overwhelming lack of labeled data needed to train state-of-the-art machine learning algorithms.However, the use of synthetic data is not without limit: in the case of imagery, training a deep learning model onpurely synthetic data typically yields poor results when applied to real world imagery. Previous work shows thatdomain adaptation", mixing real-world and synthetic data, improves performance on a target dataset. In thispaper, we train a deep neural network with synthetic imagery, including balls, cars, and overhead ship imagery,and investigate a variety of methods to adapt our model to corresponding datasets of real images.
展开▼