Single View Corridor Reconstruction

We present an autonomous algorithm for 3d reconstruction fr om a single image of an indoor scene. Most work on 3d reconstruction from vision uses multiple ima ges (for example, binocular vision/stereopsis), and recovers depths using triangulation. However, the rang e t which binocular vision is accurate is limited by the “baseline” distance between the two cameras. In c ontrast, humans are often able to look at a monocular (single-camera) image, and guess how far away obj ects were from this camera. Humans do this by bringing to bear on the problem significant prior know ledge about the world, which is used to resolve the ambiguities intrinsic to monocular vision. Foc using on the problem of 3d reconstruction in indoor scenes, in this report we present an algorithm for dep th recovery based on precise segmentation of the floor in an image. Our approach encodes prior knowledge about the environment such that the floor is flat and that walls are vertical, and is based on recogn izing the floor and the wall regions in the image. We also present experimental resulting showing our a lgorithm giving good 3d reconstruction performance using monocular images of indoor scenes.