Camera pose relative to overhead image

A computer comprising a processor and a memory, the memory storing instructions executable by the processor to:

• generate an observed overhead feature map from an overhead image of a geographic area;
• generate a ground-view feature map from a ground-view image captured by a camera within the geographic area, the camera oriented at least partially horizontally while capturing the ground-view image;
• project the ground-view feature map to an overhead view based on a geometric relationship, resulting in a projected overhead feature map;
• determine a synthesized overhead feature map based on a comparison of the ground-view feature map and the projected overhead feature map, wherein the comparison of the ground-view feature map and the projected overhead feature map includes a multi-head cross attention layer;
• determine a rotation of the camera relative to the overhead image based on the projected overhead feature map and the observed overhead feature map;
• and
• determine a translation of the camera relative to the overhead image based on the projected overhead feature map and the observed overhead feature map.