Efficient Visualization of Architectural Models from a Structure and Motion Pipeline

Abstract

State of the art three dimensional reconstruction pipelines can nowadays produce models up to several million polygons without any human intervention from a set of digital images or video. Such models are able to stretch the rendering capabilities of current hardware. We propose to augment a typical structure from motion pipeline with two additional steps, automatic fitting of high-level solid primitives and relief maps extraction, thus recovering both the overall structure of a building and its fine geometry. This not only gives birth to a more tractable and semantic model of the imaged scene, but allows for efficient and compelling rendering. We substantiate our claims showing a complete example of the described system.