MS-PS: A Multi-Scale Network for Photometric Stereo With a New Comprehensive Training Dataset - Archive ouverte HAL Access content directly
Conference Papers Year : 2023

MS-PS: A Multi-Scale Network for Photometric Stereo With a New Comprehensive Training Dataset

Abstract

The photometric stereo (PS) problem consists in reconstructing the 3D-surface of an object, thanks to a set of photographs taken under different lighting directions. In this paper, we propose a multi-scale architecture for PS which, combined with a new dataset, yields state-of-the-art results. Our proposed architecture is flexible: it permits to consider a variable number of images as well as variable image size without loss of performance. In addition, we define a set of constraints to allow the generation of a relevant synthetic dataset to train convolutional neural networks for the PS problem. Our proposed dataset is much larger than pre-existing ones, and contains many objects with challenging materials having anisotropic reflectance (e.g. metals, glass). We show on publicly available benchmarks that the combination of both these contributions drastically improves the accuracy of the estimated normal field, in comparison with previous state-of-the-art methods
Fichier principal
Vignette du fichier
F43-full.pdf (2.46 Mo) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-03863690 , version 1 (24-11-2022)
hal-03863690 , version 2 (22-08-2023)
hal-03863690 , version 3 (05-10-2023)

Licence

Attribution

Identifiers

Cite

Clément Hardy, Yvain Quéau, David Tschumperlé. MS-PS: A Multi-Scale Network for Photometric Stereo With a New Comprehensive Training Dataset. International Conference on Computer Graphics, Visualization and Computer Vision, 2023, Pilsen, Czech Republic. pp.194-203, ⟨10.24132/CSRN.3301.23⟩. ⟨hal-03863690v3⟩
205 View
111 Download

Altmetric

Share

Gmail Facebook X LinkedIn More