Learning prediction-based indoor layout estimation method and system

Abstract

The present invention discloses a learning prediction-based indoor layout estimation method and system. The method comprises the following steps that: a training set is constructed, and training samples in the training set are utilized to train a deconvolution network, wherein the training samples are room layout maps and edge maps corresponding to the room layout maps, and the room layout maps and the edge maps corresponding to the room layout maps are adopted as the input and output of the deconvolution network; a to-be-tested room layout map is inputted into the trained deconvolution network, and a predicted edge map is outputted; vanishing points in a preset direction, of the to-be-tested room layout map, are calculated, so that a plurality of sectors are generated; and sectors with local maximum edge strength are selected from the plurality of generated sectors as sampling sectors on the basis of the predicted edge map; the sampling sectors are sampled, so that a series of candidate room layout estimated maps are obtained; and a room layout map which is most similar to the predicted edge map is selected from the candidate room layout estimated maps as a final room layout map according to the similarity of the obtained room layout estimated maps and the obtained edge map.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to an indoor layout estimation method and system based on learning prediction. Background technique [0002] Room layout estimation is to simulate the room space with a best-fit 3D volumetric structure. In other words, the problem is equivalent to finding all wall-floor, wall-wall, wall-ceiling boundaries from a room. Unfortunately, these borders are mixed with various edges that are not rooms and are not always visible. Accurate room layout estimation requires computers to understand the room from a global perspective, rather than relying solely on local cues. [0003] Image features based on region information are widely used in previous works. Hedau et al. proposed a classical framework for room layout estimation: in the candidate generation stage, three mutually orthogonal vanishing points are estimated. Then by uniformly sampling from the vertical and infinite horizontal van...

AI Technical Summary

Problems solved by technology

The above methods all rely on traditional manual extraction features, there are many parameters that need to be manually adjusted, and the final estimation results have a large error

Images