Spatial Morphing Kernel Regression For Feature Interpolation.

In recent years, geotagged social media has become popular as a novel sourcefor geographic knowledge discovery. Ground-level images and videos provide adifferent perspective than overhead imagery and can be applied to a range ofapplications such as land use mapping, activity detection, pollution mapping,etc. The sparse and uneven distribution of this data presents a problem,however, for generating dense maps. We therefore investigate the problem ofspatially interpolating the high-dimensional features extracted from sparsesocial media to enable dense labeling using standard classifiers. Further, weshow how prior knowledge about region boundaries can be used to improve theinterpolation through spatial morphing kernel regression. We show that aninterpolate-then-classify framework can produce dense maps from sparseobservations but that care must be taken in choosing the interpolation method.We also show that the spatial morphing kernel improves the results.