What Is It Like Down There? Generating Dense Ground-Level Views and Image Features From Overhead Imagery Using Conditional Generative Adversarial Networks
This addresses a fundamental limitation in geographic discovery for applications relying on ground-level imagery, though it is incremental as it builds on existing cGAN methods.
The paper tackles the problem of sparse and uneven geotagged media for geographic discovery by using conditional generative adversarial networks (cGANs) to generate dense ground-level views from overhead imagery, showing that the generated images are natural-looking and that dense feature maps from this framework improve land-cover classification effectiveness compared to interpolating sparse features.
This paper investigates conditional generative adversarial networks (cGANs) to overcome a fundamental limitation of using geotagged media for geographic discovery, namely its sparse and uneven spatial distribution. We train a cGAN to generate ground-level views of a location given overhead imagery. We show the "fake" ground-level images are natural looking and are structurally similar to the real images. More significantly, we show the generated images are representative of the locations and that the representations learned by the cGANs are informative. In particular, we show that dense feature maps generated using our framework are more effective for land-cover classification than approaches which spatially interpolate features extracted from sparse ground-level images. To our knowledge, ours is the first work to use cGANs to generate ground-level views given overhead imagery and to explore the benefits of the learned representations.