Performance evaluation of time-of-flight and structured light depth sensors in radiometric/geometric variations

Hyunjung Shim, Seungkyu Lee

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Time-of-flight (ToF) and structured light depth cameras capture dense three-dimensional (3-D) geometry that is of great benefit for many computer vision problems. For the past couple of years, depth image based gesture recognition, 3-D reconstruction, and robot localization have received explosive interest in the literature. However, depth measurements present unique systematic errors, specifically when objects have specularity or translucency. We present a quantitative evaluation and analysis of depth errors using both ToF and structured light depth cameras. The evaluation framework used includes a dataset of carefully taken depth images with radiometric/geometric variations of real world objects and their ground truth depth. Our analysis and experiments reveal the different characteristics of the two sensor types and indicate that obtaining high quality depth image from real-world scene still remains a challenging, unsolved problem.

Original languageEnglish
Article number094401
JournalOptical Engineering
Volume51
Issue number9
DOIs
Publication statusPublished - 2012 Sep 1

Fingerprint

Cameras
Gesture recognition
evaluation
Systematic errors
sensors
Sensors
Computer vision
Robots
Geometry
cameras
Experiments
depth measurement
ground truth
computer vision
robots
systematic errors
geometry

All Science Journal Classification (ASJC) codes

  • Atomic and Molecular Physics, and Optics
  • Engineering(all)

Cite this

@article{472e716dbc3349918bee94c9b359a7e6,
title = "Performance evaluation of time-of-flight and structured light depth sensors in radiometric/geometric variations",
abstract = "Time-of-flight (ToF) and structured light depth cameras capture dense three-dimensional (3-D) geometry that is of great benefit for many computer vision problems. For the past couple of years, depth image based gesture recognition, 3-D reconstruction, and robot localization have received explosive interest in the literature. However, depth measurements present unique systematic errors, specifically when objects have specularity or translucency. We present a quantitative evaluation and analysis of depth errors using both ToF and structured light depth cameras. The evaluation framework used includes a dataset of carefully taken depth images with radiometric/geometric variations of real world objects and their ground truth depth. Our analysis and experiments reveal the different characteristics of the two sensor types and indicate that obtaining high quality depth image from real-world scene still remains a challenging, unsolved problem.",
author = "Hyunjung Shim and Seungkyu Lee",
year = "2012",
month = "9",
day = "1",
doi = "10.1117/1.OE.51.9.094401",
language = "English",
volume = "51",
journal = "Optical Engineering",
issn = "0091-3286",
publisher = "SPIE",
number = "9",

}

Performance evaluation of time-of-flight and structured light depth sensors in radiometric/geometric variations. / Shim, Hyunjung; Lee, Seungkyu.

In: Optical Engineering, Vol. 51, No. 9, 094401, 01.09.2012.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Performance evaluation of time-of-flight and structured light depth sensors in radiometric/geometric variations

AU - Shim, Hyunjung

AU - Lee, Seungkyu

PY - 2012/9/1

Y1 - 2012/9/1

N2 - Time-of-flight (ToF) and structured light depth cameras capture dense three-dimensional (3-D) geometry that is of great benefit for many computer vision problems. For the past couple of years, depth image based gesture recognition, 3-D reconstruction, and robot localization have received explosive interest in the literature. However, depth measurements present unique systematic errors, specifically when objects have specularity or translucency. We present a quantitative evaluation and analysis of depth errors using both ToF and structured light depth cameras. The evaluation framework used includes a dataset of carefully taken depth images with radiometric/geometric variations of real world objects and their ground truth depth. Our analysis and experiments reveal the different characteristics of the two sensor types and indicate that obtaining high quality depth image from real-world scene still remains a challenging, unsolved problem.

AB - Time-of-flight (ToF) and structured light depth cameras capture dense three-dimensional (3-D) geometry that is of great benefit for many computer vision problems. For the past couple of years, depth image based gesture recognition, 3-D reconstruction, and robot localization have received explosive interest in the literature. However, depth measurements present unique systematic errors, specifically when objects have specularity or translucency. We present a quantitative evaluation and analysis of depth errors using both ToF and structured light depth cameras. The evaluation framework used includes a dataset of carefully taken depth images with radiometric/geometric variations of real world objects and their ground truth depth. Our analysis and experiments reveal the different characteristics of the two sensor types and indicate that obtaining high quality depth image from real-world scene still remains a challenging, unsolved problem.

UR - http://www.scopus.com/inward/record.url?scp=84870583691&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84870583691&partnerID=8YFLogxK

U2 - 10.1117/1.OE.51.9.094401

DO - 10.1117/1.OE.51.9.094401

M3 - Article

VL - 51

JO - Optical Engineering

JF - Optical Engineering

SN - 0091-3286

IS - 9

M1 - 094401

ER -