Blending between street view and earth view

In one aspect, computing device(s) may determine a plurality of fragments for a three-dimensional (3D) model of a geographical location. Each fragment of the plurality of fragments may correspond to a pixel of a blended image and each fragment has a fragment color from the 3D model. The one or more computing devices may determine geospatial location data for each fragment based at least in part on latitude information, longitude information, and altitude information associated with the 3D model. For each fragment of the plurality of fragments, the one or more computing devices may identify a pixel color and an image based at least in part on the geospatial location data, determine a blending ratio based on at least one of a position and an orientation of a virtual camera, and generate the blended image based on at least the blending ratio, the pixel color, and the fragment color.