hh.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Karlsson, Stefan
Alternative names
Publications (10 of 12) Show all publications
Ranftl, A., Alonso-Fernandez, F., Karlsson, S. & Bigun, J. (2017). A Real-Time AdaBoost Cascade Face Tracker Based on Likelihood Map and Optical Flow. IET Biometrics, 6(6), 468-477
Open this publication in new window or tab >>A Real-Time AdaBoost Cascade Face Tracker Based on Likelihood Map and Optical Flow
2017 (English)In: IET Biometrics, ISSN 2047-4938, E-ISSN 2047-4946, Vol. 6, no 6, p. 468-477Article in journal (Refereed) Published
Abstract [en]

We present a novel face tracking approach where optical flow information is incorporated into a modified version of the Viola-Jones detection algorithm. In the original algorithm, detection is static, as information from previous frames is not considered; in addition, candidate windows have to pass all stages of the classification cascade, otherwise they are discarded as containing no face. In contrast, the proposed tracker preserves information about the number of classification stages passed by each window. Such information is used to build a likelihood map, which represents the probability of having a face located at that position. Tracking capabilities are provided by extrapolating the position of the likelihood map to the next frame by optical flow computation. The proposed algorithm works in real time on a standard laptop. The system is verified on the Boston Head Tracking Database, showing that the proposed algorithm outperforms the standard Viola-Jones detector in terms of detection rate and stability of the output bounding box, as well as including the capability to deal with occlusions. We also evaluate two recently published face detectors based on Convolutional Networks and Deformable Part Models, with our algorithm showing a comparable accuracy at a fraction of the computation time.

Place, publisher, year, edition, pages
Stevenage: The Institution of Engineering and Technology, 2017
National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-33836 (URN)10.1049/iet-bmt.2016.0202 (DOI)000415218200012 ()
Projects
SIDUS-AIR
Funder
Swedish Research Council, 2012-4313Knowledge Foundation, CAISRKnowledge Foundation, SIDUS-AIR
Available from: 2017-05-11 Created: 2017-05-11 Last updated: 2017-11-29Bibliographically approved
Ranftl, A., Alonso-Fernandez, F. & Karlsson, S. (2015). Face Tracking Using Optical Flow: Development of a Real-Time AdaBoost Cascade Face Tracker. In: : . Paper presented at 14th International Conference of the Biometrics Special Interest Group, BIOSIG, Darmstadt, Germany, 9-11 September, 2015.
Open this publication in new window or tab >>Face Tracking Using Optical Flow: Development of a Real-Time AdaBoost Cascade Face Tracker
2015 (English)Conference paper, Published paper (Refereed)
Abstract [en]

In this paper a novel face tracking approach is presented where optical flow information is incorporated into the Viola-Jones face detection algorithm. In the original algorithm from Viola and Jones face detection is static as information from previous frames is not considered. In contrast to the Viola-Jones face detector and also to other known dynamic enhancements, the proposed facetracker preserves information about near-positives. The algorithm builds a likelihood map from the intermediate results of the Viola-Jones algorithm which is extrapolated using optical flow. The objects get extracted from the likelihood map using image segmentation techniques. All steps can be computed very efficiently in real-time. The tracker is verified on the Boston Head Tracking Database showing that the proposed algorithm outperforms the standard Viola-Jones face detector.

Keywords
Face Tracking, Likelihood Map, Optical Flow, AdaBoost Cascade Classifier
National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-29090 (URN)10.1109/BIOSIG.2015.7314604 (DOI)000380513800012 ()2-s2.0-84959557956 (Scopus ID)
Conference
14th International Conference of the Biometrics Special Interest Group, BIOSIG, Darmstadt, Germany, 9-11 September, 2015
Funder
Swedish Research Council
Note

This paper follows a master thesis which was written within the double degree master program in Embedded and Intelligent Systems of Salzburg University of Applied Sciences, Austria and Halmstad University, Sweden.

Available from: 2015-08-06 Created: 2015-08-06 Last updated: 2018-03-22Bibliographically approved
Cooney, M. & Karlsson, S. M. (2015). Impressions of Size-Changing in a Companion Robot. In: Hugo Plácido da Silva, Pierre Chauvet, Andreas Holzinger, Stephen Fairclough & Dennis Majoe (Ed.), PhyCS 2015 – 2nd International Conference on Physiological Computing Systems, Proceedings: . Paper presented at 2nd International Conference on Physiological Computing Systems, PhyCS 2015, ESEO, Angers, Loire Valley, France, 11-13 February, 2015 (pp. 118-123). SciTePress
Open this publication in new window or tab >>Impressions of Size-Changing in a Companion Robot
2015 (English)In: PhyCS 2015 – 2nd International Conference on Physiological Computing Systems, Proceedings / [ed] Hugo Plácido da Silva, Pierre Chauvet, Andreas Holzinger, Stephen Fairclough & Dennis Majoe, SciTePress, 2015, p. 118-123Conference paper, Published paper (Refereed)
Abstract [en]

Physiological data such as head movements can be used to intuitively control a companion robot to perform useful tasks. We believe that some tasks such as reaching for high objects or getting out of a person’s way could be accomplished via size changes, but such motions should not seem threatening or bothersome. To gain insight into how size changes are perceived, the Think Aloud Method was used to gather typical impressions of a new robotic prototype which can expand in height or width based on a user’s head movements. The results indicate promise for such systems, also highlighting some potential pitfalls.

Place, publisher, year, edition, pages
SciTePress, 2015
Keywords
Adaptive Interfaces, Physiology-driven Robotics, Human-Robot Interaction, Size-Changing
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:hh:diva-29320 (URN)10.5220/0005328801180123 (DOI)2-s2.0-84938811552 (Scopus ID)978-989-758-085-7 (ISBN)
Conference
2nd International Conference on Physiological Computing Systems, PhyCS 2015, ESEO, Angers, Loire Valley, France, 11-13 February, 2015
Funder
Knowledge Foundation
Available from: 2015-08-31 Created: 2015-08-31 Last updated: 2018-03-22Bibliographically approved
Karlsson, S. (2013). Real-Time optical flow.
Open this publication in new window or tab >>Real-Time optical flow
2013 (English)Other (Other academic)
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:hh:diva-26580 (URN)
Funder
Swedish Research Council
Available from: 2014-09-24 Created: 2014-09-24 Last updated: 2018-03-22Bibliographically approved
Karlsson, S. M. & Bigun, J. (2012). Lip-motion events analysis and lip segmentation using optical flow. In: : . Paper presented at 2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2012, 16-21 June, 2012, Rhode Island, USA (pp. 138-145). Piscataway, N.J.: IEEE Press, Article ID 6239228.
Open this publication in new window or tab >>Lip-motion events analysis and lip segmentation using optical flow
2012 (English)Conference paper, Published paper (Refereed)
Abstract [en]

We propose an algorithm for detecting the mouth events of opening and closing. Our method is translation and ro- tation invariant, works at very fast speeds, and does not re- quire segmented lips. The approach is based on a recently developed optical flow algorithm that handles the motion of linear structure in a stable and consistent way.Furthermore, we provide a semi-automatic tool for gen- erating groundtruth segmentation of video data, also based on the optical flow algorithm used for tracking keypoints at faster than 200 frames/second. We provide groundtruth for 50 sessions of speech of the XM2VTS database [16] avail- able for download, and the means to segment further ses- sions at a relatively small amount of user interaction.We use the generated groundtruth to test the proposed al- gorithm for detecting events, and show it to yield promising result. The semi-automatic tool will be a useful resource for researchers in need of groundtruth segmentation from video for the XM2VTS database and others.

Place, publisher, year, edition, pages
Piscataway, N.J.: IEEE Press, 2012
Series
IEEE Computer Society Conference on Computer Vision and Pattern Recognition workshops. Proceedings, ISSN 2160-7508
Keywords
Keypoints, Linear structures, Lip segmentation, Optical flow algorithm, Rotation invariant, Semi-automatic tools, User interaction, Video data
National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-19645 (URN)10.1109/CVPRW.2012.6239228 (DOI)2-s2.0-84864974582 (Scopus ID)978-1-4673-1612-5 (ISBN)
Conference
2012 IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2012, 16-21 June, 2012, Rhode Island, USA
Funder
Swedish Research Council
Note

©2012 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Available from: 2012-10-04 Created: 2012-09-17 Last updated: 2018-03-22Bibliographically approved
Bigun, J. & Karlsson, S. M. (2011). Histogram of directions by the structure tensor. In: Proceedings of the 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies. Paper presented at The 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies, ISABEL’11, October 26-29, Barcelona. New York, NY: Association for Computing Machinery (ACM)
Open this publication in new window or tab >>Histogram of directions by the structure tensor
2011 (English)In: Proceedings of the 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies, New York, NY: Association for Computing Machinery (ACM), 2011Conference paper, Published paper (Refereed)
Abstract [en]

Many low-level features, as well as varying methods of extraction and interpretation rely on directionality analysis (for example the Hough transform, Gabor filters, SIFT descriptors and the structure tensor). The theory of the gradient based structure tensor (a.k.a. the second moment matrix) is a very well suited theoretical platform in which to analyze and explain the similarities and connections (indeed often equivalence) of supposedly different methods and features that deal with image directionality. Of special interest to this study is the SIFT descriptors (histogram of oriented gradients, HOGs). Our analysis of interrelationships of prominent directionality analysis tools offers the possibility of computation of HOGs without binning, in an algorithm of comparative time complexity.

Place, publisher, year, edition, pages
New York, NY: Association for Computing Machinery (ACM), 2011
Keywords
complex weighting, histogram of oriented gradients, structure tensor
National Category
Signal Processing
Identifiers
urn:nbn:se:hh:diva-19888 (URN)10.1145/2093698.2093794 (DOI)2-s2.0-84856697701 (Scopus ID)978-1-4503-0913-4 (ISBN)
Conference
The 4th International Symposium on Applied Sciences in Biomedical and Communication Technologies, ISABEL’11, October 26-29, Barcelona
Available from: 2012-12-19 Created: 2012-10-29 Last updated: 2018-03-22Bibliographically approved
Karlsson, S. & Bigun, J. (2011). Synthesis and detection of log-spiral codes. Paper presented at SSBA Symposium i bildanalys, 17-18 mars 2011, Linköping.
Open this publication in new window or tab >>Synthesis and detection of log-spiral codes
2011 (English)Conference paper, Published paper (Other academic)
Abstract [en]

Simultaneous positioning and identifying objects accurately and reliably is a fundamental problem in computer vision. General solutions to this problem is still challenging. For certain applications to achieve high accuracy and reliability in both tasks can be achieved if the objects can be labeled, e.g. multiple simultaneous robot tracking and navigation. We suggest a labeling technique using spiral patterns for optimal position estimation and identity recognition using the generalized structure tensor and tresholds. The technique adapts the synthesis of the labels to the frequency characteristics of the detection method. The approach has been implemented and tested by an over-head camera to track and control 8 robots in real-time.

Keywords
computer vision, robots, real-time
National Category
Computer Vision and Robotics (Autonomous Systems)
Identifiers
urn:nbn:se:hh:diva-16123 (URN)
Conference
SSBA Symposium i bildanalys, 17-18 mars 2011, Linköping
Available from: 2011-09-06 Created: 2011-09-05 Last updated: 2018-03-22Bibliographically approved
Karlsson, S. M., Pont, S. C., Koenderink, J. J. & Zisserman, A. (2010). Illuminance Flow Estimation by Regression. International Journal of Computer Vision, 90(3), 304-312
Open this publication in new window or tab >>Illuminance Flow Estimation by Regression
2010 (English)In: International Journal of Computer Vision, ISSN 0920-5691, E-ISSN 1573-1405, Vol. 90, no 3, p. 304-312Article in journal (Refereed) Published
Abstract [en]

We investigate the estimation of illuminance flow using Histograms of Oriented Gradient features (HOGs). In a regression setting, we found for both ridge regression and support vector machines, that the optimal solution shows close resemblance to the gradient based structure tensor (also known as the second moment matrix). Theoretical results are presented showing in detail how the structure tensor and the HOGs are connected. This relation will benefit computer vision tasks such as affine invariant texture/object matching using HOGs. Several properties of HOGs are presented, among others, how many bins are required for a directionality measure, and how to estimate HOGs through spatial averaging that requires no binning.

Place, publisher, year, edition, pages
Amsterdam: Elsevier, 2010
Keywords
Illuminance flow, Surface 3D texture, Histogram of oriented gradients, Illuminant estimation
National Category
Engineering and Technology
Identifiers
urn:nbn:se:hh:diva-18894 (URN)10.1007/s11263-010-0353-7 (DOI)000282782700003 ()2-s2.0-78149283686 (Scopus ID)
Available from: 2012-06-27 Created: 2012-06-25 Last updated: 2018-03-22Bibliographically approved
Karlsson, S. & Bigun, J. (2007). Multiscale complex moments of the local power spectrum. Journal of the Optical Society of America A, 24(3), 618-625
Open this publication in new window or tab >>Multiscale complex moments of the local power spectrum
2007 (English)In: Journal of the Optical Society of America A, ISSN 0740-3232, Vol. 24, no 3, p. 618-625Article in journal (Refereed) Published
Abstract [en]

Complex moments of the local power spectrum (CMP) are investigated in a multiscale context. The multiscale CMPs are shown to approximate well the 1D angular Fourier transform of the band in question. This observation is used to derive further properties of the power spectrum in terms of texture orientations or n-folded symmetry patterns. A method is presented to approximate the power spectrum using only separable filtering in the spatial domain. Interesting implications to the Gabor decomposition are shown. The number of orientations in the filter bank is related to the order of n-folded symmetry detectable. Furthermore, the multiscale CMPs can be estimated incrementally in the spatial domain, which is both fast and reliable. Experiments on power spectrum estimation, orientation estimation, and texture segmentation are presented.

Place, publisher, year, edition, pages
Washington: Optical Society of America, 2007
Keywords
Image analysis, Pattern recognition, Rotation-invariant pattern recognition
National Category
Engineering and Technology
Identifiers
urn:nbn:se:hh:diva-2069 (URN)10.1364/JOSAA.24.000618 (DOI)000244407300006 ()17301851 (PubMedID)2-s2.0-34047124800 (Scopus ID)2082/2464 (Local ID)2082/2464 (Archive number)2082/2464 (OAI)
Available from: 2008-10-20 Created: 2008-10-20 Last updated: 2018-03-23Bibliographically approved
Hoang, H., Buttazzo, G., Jonsson, M. & Karlsson, S. (2006). Computing the minimum EDF feasible deadline in periodic systems. In: 12th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, 2006. Proceedings. Paper presented at 12th IEEE Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA 2006), 16-18 August 2006, Sydney, Australia (pp. 125-134). Los Alamitos, Calif.: IEEE Press
Open this publication in new window or tab >>Computing the minimum EDF feasible deadline in periodic systems
2006 (English)In: 12th IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, 2006. Proceedings, Los Alamitos, Calif.: IEEE Press, 2006, p. 125-134Conference paper, Published paper (Refereed)
Abstract [en]

In most real-time applications, deadlines are artifices that need to be enforced to meet different performance requirements. For example, in periodic task sets, jitter requirements can be met by assigning suitable relative deadlines and guaranteeing the feasibility of the schedule. This paper presents a method (called minD) for calculating the minimum EDF-feasible deadline of a real-time task. More precisely, given a set of periodic tasks with hard real-time requirements, which is feasible under EDF, the proposed algorithm allows computing the shortest deadline that can be assigned to an arbitrary task in the set, or to a new incoming task (periodic or aperiodic), still preserving the EDF feasibility of the new task set. The algorithm has a pseudo polynomial complexity and handles arbitrary relative deadlines, which can be less than, equal to, or greater than periods.

Place, publisher, year, edition, pages
Los Alamitos, Calif.: IEEE Press, 2006
Series
IEEE International Conference on Embedded and Real-Time Computing Systems and Applications, ISSN 1533-2306 ; 2006
Keywords
Real-time, Scheduling
National Category
Computer and Information Sciences
Identifiers
urn:nbn:se:hh:diva-372 (URN)10.1109/RTCSA.2006.22 (DOI)000240851900019 ()2-s2.0-34547275304 (Scopus ID)2082/696 (Local ID)0-7695-2676-4 (ISBN)2082/696 (Archive number)2082/696 (OAI)
Conference
12th IEEE Conference on Embedded and Real-Time Computing Systems and Applications (RTCSA 2006), 16-18 August 2006, Sydney, Australia
Note

©2006 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE.

Available from: 2007-01-05 Created: 2007-01-05 Last updated: 2018-03-23Bibliographically approved
Organisations

Search in DiVA

Show all publications