Skip to Main content Skip to Navigation
Book sections

Improving thin structures in surface reconstruction from sparse point cloud

Abstract : Methods were proposed to estimate a surface from a sparse cloud of points reconstructed from images. These methods are interesting in several contexts including large scale scenes, limited computational resources and initialization of dense stereo. However they are deficient in presence of thin structures such as posts, which are often present in both urban and natural scenes: these scene components can be partly or even completely removed. Here we reduce this problem by introducing a pre-processing, assuming that (1) some of the points form polygonal chains approximating curves and occluding contours of the scene and (2) the direction of the thin structures is roughly known (e.g. vertical). In the experiments, our pre-processing improves the results of two different surface reconstruction methods applied on videos taken by helmet-held 360 cameras.
Complete list of metadatas

Cited literature [28 references]  Display  Hide  Download

https://hal.uca.fr/hal-01915349
Contributor : Pascale Dugat <>
Submitted on : Monday, October 21, 2019 - 6:56:58 PM
Last modification on : Wednesday, October 23, 2019 - 4:12:27 PM
Long-term archiving on: : Wednesday, January 22, 2020 - 9:16:34 PM

File

W17P10.pdf
Files produced by the author(s)

Identifiers

Citation

Maxime Lhuillier. Improving thin structures in surface reconstruction from sparse point cloud. ECCV 2018 workshops, 11129, Springer, pp.443-458, 2019, Lecture notes in computer science, ⟨10.1007/978-3-030-11009-3_27⟩. ⟨hal-01915349⟩

Share

Metrics

Record views

117

Files downloads

122