Translating Images into Maps
Avishkar Saha,Oscar Mendez,Chris Russell,Richard Bowden,Avishkar Saha,Oscar Mendez,Chris Russell,Richard Bowden
We approach instantaneous mapping, converting images to a top-down view of the world, as a translation problem. We show how a novel form of transformer network can be used to map from images and video directly to an overhead map or bird's-eye-view (BEV) of the world, in a single end-to-end network. We assume a 1–1 correspondence between a vertical scanline in the image, and rays passing through th...
