Removing 3D point cloud occlusion artifacts with generative adversarial networks

Visa fullständig post



Permalänk

http://urn.fi/URN:NBN:fi:hulib-202003241636
Titel: Removing 3D point cloud occlusion artifacts with generative adversarial networks
Författare: Väänänen, Pekka
Medarbetare: Helsingfors universitet, Matematisk-naturvetenskapliga fakulteten
Utgivare: Helsingin yliopisto
Datum: 2019
Språk: eng
Permanenta länken (URI): http://urn.fi/URN:NBN:fi:hulib-202003241636
http://hdl.handle.net/10138/313581
Nivå: pro gradu-avhandlingar
Ämne: Tietojenkäsittelytiede
Abstrakt: Real-world locations can be reconstructed as digital 3D models using 3D scanning. The scans often suffer from missing surface regions caused by occlusions and poor scanning geometry, limiting their usefulness for many tasks. We present an automated system that repairs small missing regions using generative adversarial networks (GANs). The system operates on heightmaps of small round surface patches distributed around the missing region. A neural network model predicts a complete plausible surface for each corrupted patch, which is then integrated to the scan. In addition to geometry, surface colors are also generated. Encouraging results are found in the color reconstruction task, but the output geometry is not clearly superior to the results of a simpler baseline spline model.


Filer under denna titel

Filer Storlek Format Granska

There are no files associated with this item.

Detta dokument registreras i samling:

Visa fullständig post