We are excited to announce that the repository will soon undergo an upgrade, featuring a new look and feel along with several enhanced features to improve your experience. Please be on the lookout for further updates and announcements regarding the launch date. We appreciate your support and look forward to unveiling the improved platform soon.
dc.contributor.author | Broekman, Andre![]() |
|
dc.contributor.author | Grabe, Petrus Johannes![]() |
|
dc.date.accessioned | 2020-10-23T10:45:41Z | |
dc.date.available | 2020-10-23T10:45:41Z | |
dc.date.issued | 2020-10 | |
dc.description | Research data for this article. PASMVS: a dataset for multi-view stereopsis training and reconstruction applications Original Data: A large collection of synthetic, path-traced renderings for use in multi-view stereopsis and 3D reconstruction applications. The material properties are primarily non-Lambertian (reflective metals). Ground truth depth maps, model geometry, object masks and camera extrinsic and intrinsic data is provided together with the rendered images. A total of 18000 samples are provided (45 camera views for 400 scenes), varying the illumination, geometry models, materials properties and camera focal length. Repository name: Mendeley Data. Data identification number: 10.17632/fhzfnwsnzf.2 URL: https://data.mendeley.com/datasets/fhzfnwsnzf/2 | en_ZA |
dc.description.abstract | A Perfectly Accurate, Synthetic dataset for Multi-View Stere- opsis (PASMVS) is presented, consisting of 400 scenes and 18,0 0 0 model renderings together with ground truth depth maps, camera intrinsic and extrinsic parameters, and binary segmentation masks. Every scene is rendered from 45 differ- ent camera views in a circular pattern, using Blender’s path- tracing rendering engine. Every scene is composed from a unique combination of two camera focal lengths, four 3D models of varying geometrical complexity, five high defini- tion, high dynamic range (HDR) environmental textures to replicate photorealistic lighting conditions and ten materials. The material properties are primarily specular, with a selec- tion of more diffuse materials for reference. The combination of highly specular and diffuse material properties increases the reconstruction ambiguity and complexity for MVS recon- struction algorithms and pipelines, and more recently, state- of-the-art architectures based on neural network implemen- tations. PASMVS serves as an addition to the wide spectrum of available image datasets employed in computer vision re- search, improving the precision required for novel research applications. | en_ZA |
dc.description.department | Civil Engineering | en_ZA |
dc.description.librarian | pm2020 | en_ZA |
dc.description.sponsorship | The Chair in Railway Engineering in the Department of Civil Engineering at the University of Pretoria. | en_ZA |
dc.description.uri | http://www.elsevier.com/locate/dib | en_ZA |
dc.identifier.citation | Broekman, A. & Gräbe, P.J. 2020, 'PASMVS : a perfectly accurate, synthetic, path-traced dataset featuring specular material properties for multi-view stereopsis training and reconstruction applications', Data in Brief, vol. 32. art. 106219, pp. 1-9. | en_ZA |
dc.identifier.issn | 2352-3409 (online) | |
dc.identifier.other | 10.1016/j.dib.2020.106219 | |
dc.identifier.uri | http://hdl.handle.net/2263/76586 | |
dc.language.iso | en | en_ZA |
dc.publisher | Elsevier | en_ZA |
dc.rights | © 2020 The Authors. Published by Elsevier Inc. This is an open access article under the CCBY license. | en_ZA |
dc.subject | Multi-view stereopsis | en_ZA |
dc.subject | 3D reconstruction | en_ZA |
dc.subject | Synthetic data | en_ZA |
dc.subject | Ground truth depth map | en_ZA |
dc.subject | Blender | en_ZA |
dc.subject | Perfectly accurate, synthetic dataset for multi-view stereopsis (PASMVS) | en_ZA |
dc.title | PASMVS : a perfectly accurate, synthetic, path-traced dataset featuring specular material properties for multi-view stereopsis training and reconstruction applications | en_ZA |
dc.type | Article | en_ZA |