Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation

Reconstructing texture-less surfaces poses unique challenges in computer vision, primarily due to the lack of specialized datasets that cater to the nuanced needs of depth and normals estimation in the absence of textural information. We introduce “Shape2.5D,” a novel, large-sc...

Full description

Saved in:
Bibliographic Details
Main Authors: Muhammad Saif Ullah Khan, Sankalp Sinha, Didier Stricker, Marcus Liwicki, Muhammad Zeshan Afzal
Format: Article
Language:English
Published: IEEE 2024-01-01
Series:IEEE Access
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10745503/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850221076761542656
author Muhammad Saif Ullah Khan
Sankalp Sinha
Didier Stricker
Marcus Liwicki
Muhammad Zeshan Afzal
author_facet Muhammad Saif Ullah Khan
Sankalp Sinha
Didier Stricker
Marcus Liwicki
Muhammad Zeshan Afzal
author_sort Muhammad Saif Ullah Khan
collection DOAJ
description Reconstructing texture-less surfaces poses unique challenges in computer vision, primarily due to the lack of specialized datasets that cater to the nuanced needs of depth and normals estimation in the absence of textural information. We introduce &#x201C;Shape2.5D,&#x201D; a novel, large-scale dataset designed to address this gap. Comprising 1.17 million frames spanning over 39,772 3D models and 48 unique objects, our dataset provides depth and surface normal maps for texture-less object reconstruction. The proposed dataset includes synthetic images rendered with 3D modeling software to simulate various lighting conditions and viewing angles. It also includes a real-world subset comprising 4,672 frames captured with a depth camera. Our comprehensive benchmarks demonstrate the dataset&#x2019;s ability to support the development of algorithms that robustly estimate depth and normals from RGB images and perform voxel reconstruction. Our open-source data generation pipeline allows the dataset to be extended and adapted for future research. The dataset is publicly available at <uri>https://github.com/saifkhichi96/Shape25D</uri>.
format Article
id doaj-art-16c0a906abd64725bf969fdcb3b43ea8
institution OA Journals
issn 2169-3536
language English
publishDate 2024-01-01
publisher IEEE
record_format Article
series IEEE Access
spelling doaj-art-16c0a906abd64725bf969fdcb3b43ea82025-08-20T02:06:50ZengIEEEIEEE Access2169-35362024-01-011217495417496410.1109/ACCESS.2024.349270310745503Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals EstimationMuhammad Saif Ullah Khan0https://orcid.org/0000-0002-7375-807XSankalp Sinha1Didier Stricker2Marcus Liwicki3https://orcid.org/0000-0003-4029-6574Muhammad Zeshan Afzal4Department of Computer Science, Rhineland-Palatinate Technical University of Kaiserslautern-Landau (RPTU), Kaiserslautern, GermanyDepartment of Computer Science, Rhineland-Palatinate Technical University of Kaiserslautern-Landau (RPTU), Kaiserslautern, GermanyDepartment of Computer Science, Rhineland-Palatinate Technical University of Kaiserslautern-Landau (RPTU), Kaiserslautern, GermanyDepartment of Computer Science, Electrical and Space Engineering, Lule&#x00E5; University of Technology, Lule&#x00E5;, SwedenDepartment of Computer Science, Rhineland-Palatinate Technical University of Kaiserslautern-Landau (RPTU), Kaiserslautern, GermanyReconstructing texture-less surfaces poses unique challenges in computer vision, primarily due to the lack of specialized datasets that cater to the nuanced needs of depth and normals estimation in the absence of textural information. We introduce &#x201C;Shape2.5D,&#x201D; a novel, large-scale dataset designed to address this gap. Comprising 1.17 million frames spanning over 39,772 3D models and 48 unique objects, our dataset provides depth and surface normal maps for texture-less object reconstruction. The proposed dataset includes synthetic images rendered with 3D modeling software to simulate various lighting conditions and viewing angles. It also includes a real-world subset comprising 4,672 frames captured with a depth camera. Our comprehensive benchmarks demonstrate the dataset&#x2019;s ability to support the development of algorithms that robustly estimate depth and normals from RGB images and perform voxel reconstruction. Our open-source data generation pipeline allows the dataset to be extended and adapted for future research. The dataset is publicly available at <uri>https://github.com/saifkhichi96/Shape25D</uri>.https://ieeexplore.ieee.org/document/10745503/Texture-less surfacesdepth estimationnormals estimation
spellingShingle Muhammad Saif Ullah Khan
Sankalp Sinha
Didier Stricker
Marcus Liwicki
Muhammad Zeshan Afzal
Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation
IEEE Access
Texture-less surfaces
depth estimation
normals estimation
title Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation
title_full Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation
title_fullStr Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation
title_full_unstemmed Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation
title_short Shape2.5D: A Dataset of Texture-Less Surfaces for Depth and Normals Estimation
title_sort shape2 5d a dataset of texture less surfaces for depth and normals estimation
topic Texture-less surfaces
depth estimation
normals estimation
url https://ieeexplore.ieee.org/document/10745503/
work_keys_str_mv AT muhammadsaifullahkhan shape25dadatasetoftexturelesssurfacesfordepthandnormalsestimation
AT sankalpsinha shape25dadatasetoftexturelesssurfacesfordepthandnormalsestimation
AT didierstricker shape25dadatasetoftexturelesssurfacesfordepthandnormalsestimation
AT marcusliwicki shape25dadatasetoftexturelesssurfacesfordepthandnormalsestimation
AT muhammadzeshanafzal shape25dadatasetoftexturelesssurfacesfordepthandnormalsestimation