Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks
This addresses the high cost of annotated image datasets for machine learning by enabling effective use of synthetic data, though it is an incremental improvement over prior domain adaptation work.
The paper tackles the problem of models trained on synthetic images failing to generalize to real images by proposing an unsupervised pixel-level domain adaptation method using GANs to transform source-domain images to appear like target-domain images, resulting in outperforming state-of-the-art methods by large margins and generalizing to unseen object classes.
Collecting well-annotated image datasets to train modern machine learning algorithms is prohibitively expensive for many tasks. One appealing alternative is rendering synthetic data where ground-truth annotations are generated automatically. Unfortunately, models trained purely on rendered images often fail to generalize to real images. To address this shortcoming, prior work introduced unsupervised domain adaptation algorithms that attempt to map representations between the two domains or learn to extract features that are domain-invariant. In this work, we present a new approach that learns, in an unsupervised manner, a transformation in the pixel space from one domain to the other. Our generative adversarial network (GAN)-based method adapts source-domain images to appear as if drawn from the target domain. Our approach not only produces plausible samples, but also outperforms the state-of-the-art on a number of unsupervised domain adaptation scenarios by large margins. Finally, we demonstrate that the adaptation process generalizes to object classes unseen during training.