Sparse depth calculation using real-time key-point: Detection and structure from motion for advanced driver assist systems

Charan D. Prakash, Jinjin Li, Farshad Akhbari, Lina Karam

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper presents a system for calculating depth using a single camera with a focus on advanced driver assist systems. The proposed system consists of an improved structure from motion (SfM) approach. First, a novel multi-scale fast feature point detector (MFFPD) is proposed for detecting keypoints in the image in real-time with high accuracy. Secondly, a method is presented for sparse depth calculation at the detected key-points locations using multi-view 3D modeling. The proposed SfM system is capable of processing multiple video frames from a single planar or fisheye camera setup and is resilient to camera calibration parameter drifts. The algorithm pipeline is implemented using OpenCV/C++. Results are presented for sets of images that contain temporal motion and sets that contain lateral motion corresponding, respectively, to views from the front and side video cameras of a car.

Original languageEnglish (US)
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages740-751
Number of pages12
Volume8887
ISBN (Print)9783319142487
StatePublished - 2014
Event10th International Symposium on Visual Computing, ISVC 2014 - Las Vegas, United States
Duration: Dec 8 2014Dec 10 2014

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume8887
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other10th International Symposium on Visual Computing, ISVC 2014
CountryUnited States
CityLas Vegas
Period12/8/1412/10/14

Fingerprint

Structure from Motion
Driver
Cameras
Real-time
Camera
Video cameras
Point Location
Camera Calibration
3D Modeling
Motion
Feature Point
Railroad cars
Pipelines
C++
Calibration
Detectors
Lateral
High Accuracy
Detector
Processing

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Prakash, C. D., Li, J., Akhbari, F., & Karam, L. (2014). Sparse depth calculation using real-time key-point: Detection and structure from motion for advanced driver assist systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 8887, pp. 740-751). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8887). Springer Verlag.

Sparse depth calculation using real-time key-point : Detection and structure from motion for advanced driver assist systems. / Prakash, Charan D.; Li, Jinjin; Akhbari, Farshad; Karam, Lina.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8887 Springer Verlag, 2014. p. 740-751 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 8887).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Prakash, CD, Li, J, Akhbari, F & Karam, L 2014, Sparse depth calculation using real-time key-point: Detection and structure from motion for advanced driver assist systems. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 8887, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 8887, Springer Verlag, pp. 740-751, 10th International Symposium on Visual Computing, ISVC 2014, Las Vegas, United States, 12/8/14.
Prakash CD, Li J, Akhbari F, Karam L. Sparse depth calculation using real-time key-point: Detection and structure from motion for advanced driver assist systems. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8887. Springer Verlag. 2014. p. 740-751. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
Prakash, Charan D. ; Li, Jinjin ; Akhbari, Farshad ; Karam, Lina. / Sparse depth calculation using real-time key-point : Detection and structure from motion for advanced driver assist systems. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 8887 Springer Verlag, 2014. pp. 740-751 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{42c55ae5c4b745fdaf3a5eecb0209060,
title = "Sparse depth calculation using real-time key-point: Detection and structure from motion for advanced driver assist systems",
abstract = "This paper presents a system for calculating depth using a single camera with a focus on advanced driver assist systems. The proposed system consists of an improved structure from motion (SfM) approach. First, a novel multi-scale fast feature point detector (MFFPD) is proposed for detecting keypoints in the image in real-time with high accuracy. Secondly, a method is presented for sparse depth calculation at the detected key-points locations using multi-view 3D modeling. The proposed SfM system is capable of processing multiple video frames from a single planar or fisheye camera setup and is resilient to camera calibration parameter drifts. The algorithm pipeline is implemented using OpenCV/C++. Results are presented for sets of images that contain temporal motion and sets that contain lateral motion corresponding, respectively, to views from the front and side video cameras of a car.",
author = "Prakash, {Charan D.} and Jinjin Li and Farshad Akhbari and Lina Karam",
year = "2014",
language = "English (US)",
isbn = "9783319142487",
volume = "8887",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "740--751",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Sparse depth calculation using real-time key-point

T2 - Detection and structure from motion for advanced driver assist systems

AU - Prakash, Charan D.

AU - Li, Jinjin

AU - Akhbari, Farshad

AU - Karam, Lina

PY - 2014

Y1 - 2014

N2 - This paper presents a system for calculating depth using a single camera with a focus on advanced driver assist systems. The proposed system consists of an improved structure from motion (SfM) approach. First, a novel multi-scale fast feature point detector (MFFPD) is proposed for detecting keypoints in the image in real-time with high accuracy. Secondly, a method is presented for sparse depth calculation at the detected key-points locations using multi-view 3D modeling. The proposed SfM system is capable of processing multiple video frames from a single planar or fisheye camera setup and is resilient to camera calibration parameter drifts. The algorithm pipeline is implemented using OpenCV/C++. Results are presented for sets of images that contain temporal motion and sets that contain lateral motion corresponding, respectively, to views from the front and side video cameras of a car.

AB - This paper presents a system for calculating depth using a single camera with a focus on advanced driver assist systems. The proposed system consists of an improved structure from motion (SfM) approach. First, a novel multi-scale fast feature point detector (MFFPD) is proposed for detecting keypoints in the image in real-time with high accuracy. Secondly, a method is presented for sparse depth calculation at the detected key-points locations using multi-view 3D modeling. The proposed SfM system is capable of processing multiple video frames from a single planar or fisheye camera setup and is resilient to camera calibration parameter drifts. The algorithm pipeline is implemented using OpenCV/C++. Results are presented for sets of images that contain temporal motion and sets that contain lateral motion corresponding, respectively, to views from the front and side video cameras of a car.

UR - http://www.scopus.com/inward/record.url?scp=84916618722&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84916618722&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9783319142487

VL - 8887

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 740

EP - 751

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -