Please use this identifier to cite or link to this item:
Title: SliceNet: deep dense depth estimation from a single indoor panorama using a slice-based representation
Authors: Pintore, Giovanni 
Agus, Marco 
Almansa Aránega, Eva María 
Schneider, Jens 
Gobbetti, Enrico 
Affiliations: CRS4 
Keywords: visual computing;indoor reconstruction;indoor panorama
Issue Date: Jun-2021
Project: Advanced Visual and Geometric Computing for 3D Capture, Display, and Fabrication 
Related Publication(s): Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)
Start page: 11536
End page: 11545
We introduce a novel deep neural network to estimate a depth map from a single monocular indoor panorama. The network directly works on the equirectangular projection, exploiting the properties of indoor 360-degree images. Starting from the fact that gravity plays an important role in the design and construction of man-made indoor scenes, we propose a compact representation of the scene into vertical slices of the sphere, and we exploit long- and short-term relationships among slices to recover the equirectangular depth map. Our design makes it possible to maintain high-resolution information in the extracted features even with a deep network. The experimental results demonstrate that our method outperforms current state-of-the-art solutions in prediction accuracy, particularly for real-world data.
Appears in Collections:CRS4 publications

Files in This Item:
File Description SizeFormat
cvpr2021-slicenet.pdfGreen OA copy4,38 MBAdobe PDFView/Open
Show full item record

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.