Benutzer: Gast  Login
Titel:

SoftPoolNet: Shape Descriptor for Point Cloud Completion and Classification

Dokumenttyp:
Konferenzbeitrag
Autor(en):
Wang, Y.; Tan, D. J.; Navab, N.; Tombari, F.
Abstract:
Point clouds are often the default choice for many applications as they exhibit more flexibility and efficiency than volumetric data. Nevertheless, their unorganized nature – points are stored in an unordered way – makes them less suited to be processed by deep learning pipelines. In this paper, we propose a method for 3D object completion and classification based on point clouds. We introduce a new way of organizing the extracted features based on their activations, which we name soft pooling....     »
Stichworte:
ECCV,CAMP,CAMPComputerVision,ComputerVision,ARXIV,CNN,Point Cloud Completion,Deep Learning
Herausgeber:
Vedaldi, Andrea; Bischof, Horst; Brox, Thomas; Frahm, Jan-Michael
Kongress- / Buchtitel:
Computer Vision -- ECCV 2020
Verlag / Institution:
Springer International Publishing
Verlagsort:
Cham
Jahr:
2020
Seiten:
70--85
Print-ISBN:
978-3-030-58580-8
 BibTeX