Smart Hypothesis Generation for Efficient and Robust Room Layout Estimation
This work addresses the need for efficient and robust room layout estimation in computer vision applications, offering improvements over slow and unreliable existing methods.
The paper tackles the problem of estimating room layout from a single RGB image by proposing a hypothesize-and-test method based on semantic segmentation, which significantly outperforms state-of-the-art approaches on three benchmark datasets.
We propose a novel method to efficiently estimate the spatial layout of a room from a single monocular RGB image. As existing approaches based on low-level feature extraction, followed by a vanishing point estimation are very slow and often unreliable in realistic scenarios, we build on semantic segmentation of the input image. To obtain better segmentations, we introduce a robust, accurate and very efficient hypothesize-and-test scheme. The key idea is to use three segmentation hypotheses, each based on a different number of visible walls. For each hypothesis, we predict the image locations of the room corners and select the hypothesis for which the layout estimated from the room corners is consistent with the segmentation. We demonstrate the efficiency and robustness of our method on three challenging benchmark datasets, where we significantly outperform the state-of-the-art.