Fast Bridgeless Pyramid Segmentation for Organized Point Clouds

D 2021

Fast Bridgeless Pyramid Segmentation for Organized Point Clouds

MADARAS, Martin, Martin STUCHLIK a Matúš TALČÍK

Základní údaje

Originální název

Fast Bridgeless Pyramid Segmentation for Organized Point Clouds

Autoři

MADARAS, Martin, Martin STUCHLIK a Matúš TALČÍK (703 Slovensko, domácí)

Vydání

SETUBAL, VISAPP: PROCEEDINGS OF THE 16TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS - VOL. 4: VISAPP, od s. 205-210, 6 s. 2021

Nakladatel

SCITEPRESS

Další údaje

Jazyk

angličtina

Typ výsledku

Stať ve sborníku

Obor

10201 Computer sciences, information science, bioinformatics

Stát vydavatele

Portugalsko

Utajení

není předmětem státního či obchodního tajemství

Forma vydání

elektronická verze "online"

Kód RIV

RIV/00216224:14330/21:00123673

Organizační jednotka

Fakulta informatiky

ISBN

978-989-758-488-6

DOI

http://dx.doi.org/10.5220/0010163802050210

UT WoS

000668577400019

Klíčová slova anglicky

Point Cloud; Segmentation; Parallel; Pyramid; GPU; CUDA

Štítky

core_B, firank_B

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 23. 5. 2022 15:09, RNDr. Pavel Šmerk, Ph.D.

Anotace

V originále

An intelligent automatic robotic system needs to understand the world as fast as possible. A common way to capture the world is to use a depth camera. The depth camera produces an organized point cloud that later needs to be processed to understand the scene. Usually, segmentation is one of the first preprocessing steps for the data processing pipeline. Our proposed pyramid segmentation is a simple, fast and lightweight split-and-merge method designed for depth cameras. The algorithm consists of two steps, edge detection and a hierarchical method for bridgeless labeling of connected components. The pyramid segmentation generates the seeds hierarchically, in a top-down manner, from the largest regions to the smallest ones. The neighboring areas around the seeds are filled in a parallel manner, by rendering axis-aligned line primitives, which makes the performance of the method fast. The hierarchical approach of labeling enables to connect neighboring segments without unnecessary bridges in a parallel way that can be efficiently implemented using CUDA.

Citovat

MADARAS, Martin, Martin STUCHLIK a Matúš TALČÍK. Fast Bridgeless Pyramid Segmentation for Organized Point Clouds. Online. In Farinella, GM Radeva, P Braz, J Bouatouch, K. VISAPP: PROCEEDINGS OF THE 16TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS - VOL. 4: VISAPP. SETUBAL: SCITEPRESS, 2021, s. 205-210. ISBN 978-989-758-488-6. Dostupné z: https://dx.doi.org/10.5220/0010163802050210.

@inproceedings{1820089,
   author = {Madaras, Martin and Stuchlik, Martin and Talčík, Matúš},
   address = {SETUBAL},
   booktitle = {VISAPP: PROCEEDINGS OF THE 16TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS - VOL. 4: VISAPP},
   doi = {http://dx.doi.org/10.5220/0010163802050210},
   editor = {Farinella, GM Radeva, P Braz, J Bouatouch, K},
   keywords = {Point Cloud; Segmentation; Parallel; Pyramid; GPU; CUDA},
   howpublished = {elektronická verze "online"},
   language = {eng},
   location = {SETUBAL},
   isbn = {978-989-758-488-6},
   pages = {205-210},
   publisher = {SCITEPRESS},
   title = {Fast Bridgeless Pyramid Segmentation for Organized Point Clouds},
   year = {2021}
}

TY  - JOUR
ID  - 1820089
AU  - Madaras, Martin - Stuchlik, Martin - Talčík, Matúš
PY  - 2021
TI  - Fast Bridgeless Pyramid Segmentation for Organized Point Clouds
PB  - SCITEPRESS
CY  - SETUBAL
SN  - 9789897584886
KW  - Point Cloud
KW  - Segmentation
KW  - Parallel
KW  - Pyramid
KW  - GPU
KW  - CUDA
N2  - An intelligent automatic robotic system needs to understand the world as fast as possible. A common way to capture the world is to use a depth camera. The depth camera produces an organized point cloud that later needs to be processed to understand the scene. Usually, segmentation is one of the first preprocessing steps for the data processing pipeline. Our proposed pyramid segmentation is a simple, fast and lightweight split-and-merge method designed for depth cameras. The algorithm consists of two steps, edge detection and a hierarchical method for bridgeless labeling of connected components. The pyramid segmentation generates the seeds hierarchically, in a top-down manner, from the largest regions to the smallest ones. The neighboring areas around the seeds are filled in a parallel manner, by rendering axis-aligned line primitives, which makes the performance of the method fast. The hierarchical approach of labeling enables to connect neighboring segments without unnecessary bridges in a parallel way that can be efficiently implemented using CUDA.
ER  -

MADARAS, Martin, Martin STUCHLIK a Matúš TALČÍK. Fast Bridgeless Pyramid Segmentation for Organized Point Clouds. Online. In Farinella, GM Radeva, P Braz, J Bouatouch, K. \textit{VISAPP: PROCEEDINGS OF THE 16TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS - VOL. 4: VISAPP}. SETUBAL: SCITEPRESS, 2021, s.~205-210. ISBN~978-989-758-488-6. Dostupné z: https://dx.doi.org/10.5220/0010163802050210.

Podrobný výpis o publikaci