Visual odometry based omni-directional hyperlapse

Prachi Rani; Arpit Jangid; Vinay P. Namboodiri; K. S. Venkatesh

doi:10.1007/978-981-13-0020-2_1

Visual odometry based omni-directional hyperlapse

Prachi Rani, Arpit Jangid, Vinay P. Namboodiri, K. S. Venkatesh

Department of Computer Science

Indian Institute of Technology Kanpur

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

4 Citations (SciVal)

Abstract

The prohibitive amounts of time required to review the large amounts of data captured by surveillance and other cameras has brought into question the very utility of large scale video logging. Yet, one recognizes that such logging and analysis are indispensable to security applications. The only way out of this paradox is to devise expedited browsing, by the creation of hyperlapse. We address the hyperlapse problem for the very challenging category of intensive egomotion which makes the hyperlapse highly jerky. We propose an economical approach for trajectory estimation based on Visual Odometry and implement cost functions to penalize pose and path deviations. Also, this is implemented on data taken by omni-directional camera, so that the viewer can opt to observe any direction while browsing. This requires many innovations, including handling the massive radial distortions and implementing scene stabilization that need to be operated upon the least distorted region of the omni view.

Original language	English
Title of host publication	Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers
Editors	Chetan Arora, Renu Rameshan, Sumantra Dutta Roy
Publisher	Springer Verlag
Pages	3-13
Number of pages	11
ISBN (Print)	9789811300196
DOIs	https://doi.org/10.1007/978-981-13-0020-2_1
Publication status	Published - 1 Jan 2018
Event	6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017 - Mandi, India Duration: 16 Dec 2017 → 19 Dec 2017

Publication series

Name	Communications in Computer and Information Science
Volume	841
ISSN (Print)	1865-0929

Conference

Conference	6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017
Country/Territory	India
City	Mandi
Period	16/12/17 → 19/12/17

Funding

Acknowledgment. This project is a part of postgraduate dissertation research work at Indian Institute of Technology Kanpur, India, financially supported by the Indian government organization Defence Research and Development Organisation.

ASJC Scopus subject areas

General Computer Science
General Mathematics

Access to Document

10.1007/978-981-13-0020-2_1

Cite this

Rani, P., Jangid, A., Namboodiri, V. P., & Venkatesh, K. S. (2018). Visual odometry based omni-directional hyperlapse. In C. Arora, R. Rameshan, & S. Dutta Roy (Eds.), Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers (pp. 3-13). (Communications in Computer and Information Science; Vol. 841). Springer Verlag. https://doi.org/10.1007/978-981-13-0020-2_1

Visual odometry based omni-directional hyperlapse. / Rani, Prachi; Jangid, Arpit; Namboodiri, Vinay P. et al.
Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers. ed. / Chetan Arora; Renu Rameshan; Sumantra Dutta Roy. Springer Verlag, 2018. p. 3-13 (Communications in Computer and Information Science; Vol. 841).

Research output: Chapter or section in a book/report/conference proceeding › Chapter in a published conference proceeding

Rani, P, Jangid, A, Namboodiri, VP & Venkatesh, KS 2018, Visual odometry based omni-directional hyperlapse. in C Arora, R Rameshan & S Dutta Roy (eds), Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers. Communications in Computer and Information Science, vol. 841, Springer Verlag, pp. 3-13, 6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017, Mandi, India, 16/12/17. https://doi.org/10.1007/978-981-13-0020-2_1

Rani P, Jangid A, Namboodiri VP, Venkatesh KS. Visual odometry based omni-directional hyperlapse. In Arora C, Rameshan R, Dutta Roy S, editors, Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers. Springer Verlag. 2018. p. 3-13. (Communications in Computer and Information Science). doi: 10.1007/978-981-13-0020-2_1

Rani, Prachi ; Jangid, Arpit ; Namboodiri, Vinay P. et al. / Visual odometry based omni-directional hyperlapse. Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers. editor / Chetan Arora ; Renu Rameshan ; Sumantra Dutta Roy. Springer Verlag, 2018. pp. 3-13 (Communications in Computer and Information Science).

@inproceedings{07b552ae46804b388438e781d6e8b5ba,

title = "Visual odometry based omni-directional hyperlapse",

abstract = "The prohibitive amounts of time required to review the large amounts of data captured by surveillance and other cameras has brought into question the very utility of large scale video logging. Yet, one recognizes that such logging and analysis are indispensable to security applications. The only way out of this paradox is to devise expedited browsing, by the creation of hyperlapse. We address the hyperlapse problem for the very challenging category of intensive egomotion which makes the hyperlapse highly jerky. We propose an economical approach for trajectory estimation based on Visual Odometry and implement cost functions to penalize pose and path deviations. Also, this is implemented on data taken by omni-directional camera, so that the viewer can opt to observe any direction while browsing. This requires many innovations, including handling the massive radial distortions and implementing scene stabilization that need to be operated upon the least distorted region of the omni view.",

author = "Prachi Rani and Arpit Jangid and Namboodiri, {Vinay P.} and Venkatesh, {K. S.}",

year = "2018",

month = jan,

day = "1",

doi = "10.1007/978-981-13-0020-2_1",

language = "English",

isbn = "9789811300196",

series = "Communications in Computer and Information Science",

publisher = "Springer Verlag",

pages = "3--13",

editor = "Chetan Arora and Renu Rameshan and {Dutta Roy}, Sumantra",

booktitle = "Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers",

address = "Germany",

note = "6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017 ; Conference date: 16-12-2017 Through 19-12-2017",

}

TY - GEN

T1 - Visual odometry based omni-directional hyperlapse

AU - Rani, Prachi

AU - Jangid, Arpit

AU - Namboodiri, Vinay P.

AU - Venkatesh, K. S.

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The prohibitive amounts of time required to review the large amounts of data captured by surveillance and other cameras has brought into question the very utility of large scale video logging. Yet, one recognizes that such logging and analysis are indispensable to security applications. The only way out of this paradox is to devise expedited browsing, by the creation of hyperlapse. We address the hyperlapse problem for the very challenging category of intensive egomotion which makes the hyperlapse highly jerky. We propose an economical approach for trajectory estimation based on Visual Odometry and implement cost functions to penalize pose and path deviations. Also, this is implemented on data taken by omni-directional camera, so that the viewer can opt to observe any direction while browsing. This requires many innovations, including handling the massive radial distortions and implementing scene stabilization that need to be operated upon the least distorted region of the omni view.

AB - The prohibitive amounts of time required to review the large amounts of data captured by surveillance and other cameras has brought into question the very utility of large scale video logging. Yet, one recognizes that such logging and analysis are indispensable to security applications. The only way out of this paradox is to devise expedited browsing, by the creation of hyperlapse. We address the hyperlapse problem for the very challenging category of intensive egomotion which makes the hyperlapse highly jerky. We propose an economical approach for trajectory estimation based on Visual Odometry and implement cost functions to penalize pose and path deviations. Also, this is implemented on data taken by omni-directional camera, so that the viewer can opt to observe any direction while browsing. This requires many innovations, including handling the massive radial distortions and implementing scene stabilization that need to be operated upon the least distorted region of the omni view.

UR - http://www.scopus.com/inward/record.url?scp=85046496679&partnerID=8YFLogxK

U2 - 10.1007/978-981-13-0020-2_1

DO - 10.1007/978-981-13-0020-2_1

M3 - Chapter in a published conference proceeding

AN - SCOPUS:85046496679

SN - 9789811300196

T3 - Communications in Computer and Information Science

SP - 3

EP - 13

BT - Computer Vision, Pattern Recognition, Image Processing, and Graphics - 6th National Conference, NCVPRIPG 2017, Revised Selected Papers

A2 - Arora, Chetan

A2 - Rameshan, Renu

A2 - Dutta Roy, Sumantra

PB - Springer Verlag

T2 - 6th National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, NCVPRIPG 2017

Y2 - 16 December 2017 through 19 December 2017

ER -

Visual odometry based omni-directional hyperlapse

Abstract

Publication series

Conference

Funding

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this