Please use this identifier to cite or link to this item:
|Title:||AtlantaNet: Inferring the 3D Indoor Layout from a Single 360 Image beyond the Manhattan World Assumption||Authors:||Pintore, Giovanni
|Keywords:||3D floor plan recovery;panoramic images;360 images;data-driven reconstruction;structured indoor reconstruction;indoor panorama;room layout estimation;holistic scene reconstruction;visual computing||Issue Date:||Nov-2020||Publisher:||Springer||Project:||Advanced Visual and Geometric Computing for 3D Capture, Display, and Fabrication
|Related Publication(s):||Computer Vision – ECCV 2020 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part VIII||Journal:||LNCS||Volume:||12353||Start page:||432||End page:||448||Conference:||ECCV 2020||Abstract:||
We introduce a novel end-to-end approach to predict a 3D room layout from a single panoramic image. Compared to recent state-of-the-art works, our method is not limited to Manhattan World environments, and can reconstruct rooms bounded by vertical walls that do not form right angles or are curved - i.e., Atlanta World models. In our approach, we project the original gravity-aligned panoramic image on two horizontal planes, one above and one below the camera. This representation encodes all the information needed to recover the Atlanta World 3D bounding surfaces of the room in the form of a 2D room footprint on the floor plan and a room height. To predict the 3D layout, we propose an encoder-decoder neural network architecture, leveraging Recurrent Neural Networks (RNNs) to capture long-range geometric patterns, and exploiting a customized training strategy based on domain-specific knowledge. The experimental results demonstrate that our method outperforms state-of-the-art solutions in prediction accuracy, in particular in cases of complex wall layouts or curved wall footprints.
|Appears in Collections:||CRS4 publications|
Show full item record
Files in This Item:
checked on Nov 11, 2020
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.