Discriminative Scale Space Tracking | IEEE Journals & Magazine | IEEE Xplore

Abstract:

Accurate scale estimation of a target is a challenging research problem in visual object tracking. Most state-of-the-art methods employ an exhaustive scale search to esti...Show More

Abstract:

Accurate scale estimation of a target is a challenging research problem in visual object tracking. Most state-of-the-art methods employ an exhaustive scale search to estimate the target size. The exhaustive search strategy is computationally expensive and struggles when encountered with large scale variations. This paper investigates the problem of accurate and robust scale estimation in a tracking-by-detection framework. We propose a novel scale adaptive tracking approach by learning separate discriminative correlation filters for translation and scale estimation. The explicit scale filter is learned online using the target appearance sampled at a set of different scales. Contrary to standard approaches, our method directly learns the appearance change induced by variations in the target scale. Additionally, we investigate strategies to reduce the computational cost of our approach. Extensive experiments are performed on the OTB and the VOT2014 datasets. Compared to the standard exhaustive scale search, our approach achieves a gain of 2.5 percent in average overlap precision on the OTB dataset. Additionally, our method is computationally efficient, operating at a 50 percent higher frame rate compared to the exhaustive scale search. Our method obtains the top rank in performance by outperforming 19 state-of-the-art trackers on OTB and 37 state-of-the-art trackers on VOT2014.
Published in: IEEE Transactions on Pattern Analysis and Machine Intelligence ( Volume: 39, Issue: 8, 01 August 2017)
Page(s): 1561 - 1575
Date of Publication: 15 September 2016

ISSN Information:

PubMed ID: 27654137

Funding Agency:

Author image of Martin Danelljan
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Martin Danelljan received the MSc degree in electrical engineering from Linköping University, in 2014. He is working toward the PhD degree in the Computer Vision Laboratory, Linköping University, Sweden. His research interests include machine learning methods for visual tracking and statistical models for point set registration. He has published several papers in major computer vision conferences, including CVPR, ICCV...Show More
Martin Danelljan received the MSc degree in electrical engineering from Linköping University, in 2014. He is working toward the PhD degree in the Computer Vision Laboratory, Linköping University, Sweden. His research interests include machine learning methods for visual tracking and statistical models for point set registration. He has published several papers in major computer vision conferences, including CVPR, ICCV...View more
Author image of Gustav Häger
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Gustav Häger is working toward the PhD degree at Linköping University. His research interests include machine learning, with particular focus on methods for online tracking and detection. He also supervises the LiU-humanoids standard platform league team for the RoboCup competitions. He is a student member of the IEEE.
Gustav Häger is working toward the PhD degree at Linköping University. His research interests include machine learning, with particular focus on methods for online tracking and detection. He also supervises the LiU-humanoids standard platform league team for the RoboCup competitions. He is a student member of the IEEE.View more
Author image of Fahad Shahbaz Khan
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Fahad Shahbaz Khan received the MSc degree in intelligent systems design from Chalmers University of Technology, Sweden, and the PhD degree in computer vision from Autonomous University of Barcelona, Spain. He is a research fellow in the Computer Vision Laboratory, Linköping University, Sweden. From 2012 to 2014, he was post doctoral fellow in the Computer Vision Laboratory, Linköping University, Sweden. His research ...Show More
Fahad Shahbaz Khan received the MSc degree in intelligent systems design from Chalmers University of Technology, Sweden, and the PhD degree in computer vision from Autonomous University of Barcelona, Spain. He is a research fellow in the Computer Vision Laboratory, Linköping University, Sweden. From 2012 to 2014, he was post doctoral fellow in the Computer Vision Laboratory, Linköping University, Sweden. His research ...View more
Author image of Michael Felsberg
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Michael Felsberg received the PhD degree in engineering from the University of Kiel, Kiel, Germany, in 2002. Since 2008, he has been a full professor and the head of the Computer Vision Laboratory, Linköping University, Linköping, Sweden. His current research interests include signal processing methods for image analysis, computer and robot vision, and machine learning. He has published more than 100 reviewed conferen...Show More
Michael Felsberg received the PhD degree in engineering from the University of Kiel, Kiel, Germany, in 2002. Since 2008, he has been a full professor and the head of the Computer Vision Laboratory, Linköping University, Linköping, Sweden. His current research interests include signal processing methods for image analysis, computer and robot vision, and machine learning. He has published more than 100 reviewed conferen...View more

Author image of Martin Danelljan
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Martin Danelljan received the MSc degree in electrical engineering from Linköping University, in 2014. He is working toward the PhD degree in the Computer Vision Laboratory, Linköping University, Sweden. His research interests include machine learning methods for visual tracking and statistical models for point set registration. He has published several papers in major computer vision conferences, including CVPR, ICCV, and ECCV. In 2014, he was awarded the Tryggve Holm medal for outstanding student achievements and grades and the Swedish Computer Society award for best master’s thesis. He has achieved the top rank in the Visual Object Tracking (VOT) Challenge 2014, the OpenCV State-of-the-Art Vision Challenge 2015 in Tracking and the VOT Thermal Infrared Challenge 2015. He is a student member of the IEEE.
Martin Danelljan received the MSc degree in electrical engineering from Linköping University, in 2014. He is working toward the PhD degree in the Computer Vision Laboratory, Linköping University, Sweden. His research interests include machine learning methods for visual tracking and statistical models for point set registration. He has published several papers in major computer vision conferences, including CVPR, ICCV, and ECCV. In 2014, he was awarded the Tryggve Holm medal for outstanding student achievements and grades and the Swedish Computer Society award for best master’s thesis. He has achieved the top rank in the Visual Object Tracking (VOT) Challenge 2014, the OpenCV State-of-the-Art Vision Challenge 2015 in Tracking and the VOT Thermal Infrared Challenge 2015. He is a student member of the IEEE.View more
Author image of Gustav Häger
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Gustav Häger is working toward the PhD degree at Linköping University. His research interests include machine learning, with particular focus on methods for online tracking and detection. He also supervises the LiU-humanoids standard platform league team for the RoboCup competitions. He is a student member of the IEEE.
Gustav Häger is working toward the PhD degree at Linköping University. His research interests include machine learning, with particular focus on methods for online tracking and detection. He also supervises the LiU-humanoids standard platform league team for the RoboCup competitions. He is a student member of the IEEE.View more
Author image of Fahad Shahbaz Khan
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Fahad Shahbaz Khan received the MSc degree in intelligent systems design from Chalmers University of Technology, Sweden, and the PhD degree in computer vision from Autonomous University of Barcelona, Spain. He is a research fellow in the Computer Vision Laboratory, Linköping University, Sweden. From 2012 to 2014, he was post doctoral fellow in the Computer Vision Laboratory, Linköping University, Sweden. His research interests include object recognition, action recognition, and visual tracking. He has published articles in high-impact computer vision journals and conferences in these areas. He is a member of the IEEE.
Fahad Shahbaz Khan received the MSc degree in intelligent systems design from Chalmers University of Technology, Sweden, and the PhD degree in computer vision from Autonomous University of Barcelona, Spain. He is a research fellow in the Computer Vision Laboratory, Linköping University, Sweden. From 2012 to 2014, he was post doctoral fellow in the Computer Vision Laboratory, Linköping University, Sweden. His research interests include object recognition, action recognition, and visual tracking. He has published articles in high-impact computer vision journals and conferences in these areas. He is a member of the IEEE.View more
Author image of Michael Felsberg
Department of Electrical Engineering, Linköping University, Linköping, Sweden
Michael Felsberg received the PhD degree in engineering from the University of Kiel, Kiel, Germany, in 2002. Since 2008, he has been a full professor and the head of the Computer Vision Laboratory, Linköping University, Linköping, Sweden. His current research interests include signal processing methods for image analysis, computer and robot vision, and machine learning. He has published more than 100 reviewed conference papers, journal articles, and book contributions. He received awards from the German Pattern Recognition Society in 2000, 2004, and 2005, from the Swedish Society for Automated Image Analysis in 2007 and 2010, from Conference on Information Fusion in 2011 (Honorable Mention), and from the CVPR Workshop on Mobile Vision 2014. He has achieved top ranks on various challenges (VOT: 3rd 2013, 1st 2014, 2nd 2015; VOT-TIR: 1st 2015; OpenCV Tracking: 1st 2015; KITTI Stereo Odometry: 1st 2015, March). He has coordinated the EU projects COSPAL and DIPLECS, he is an associate editor of the Journal of Mathematical Imaging and Vision, the Journal of Image and Vision Computing, the Journal of Real-Time Image Processing, and the Frontiers in Robotics and AI. He was a publication chair of the International Conference on Pattern Recognition 2014 and Track Chair 2016, he was the general co-chair of the DAGM symposium in 2011, and he will be a general chair of CAIP 2017. He is a senior member of the IEEE.
Michael Felsberg received the PhD degree in engineering from the University of Kiel, Kiel, Germany, in 2002. Since 2008, he has been a full professor and the head of the Computer Vision Laboratory, Linköping University, Linköping, Sweden. His current research interests include signal processing methods for image analysis, computer and robot vision, and machine learning. He has published more than 100 reviewed conference papers, journal articles, and book contributions. He received awards from the German Pattern Recognition Society in 2000, 2004, and 2005, from the Swedish Society for Automated Image Analysis in 2007 and 2010, from Conference on Information Fusion in 2011 (Honorable Mention), and from the CVPR Workshop on Mobile Vision 2014. He has achieved top ranks on various challenges (VOT: 3rd 2013, 1st 2014, 2nd 2015; VOT-TIR: 1st 2015; OpenCV Tracking: 1st 2015; KITTI Stereo Odometry: 1st 2015, March). He has coordinated the EU projects COSPAL and DIPLECS, he is an associate editor of the Journal of Mathematical Imaging and Vision, the Journal of Image and Vision Computing, the Journal of Real-Time Image Processing, and the Frontiers in Robotics and AI. He was a publication chair of the International Conference on Pattern Recognition 2014 and Track Chair 2016, he was the general co-chair of the DAGM symposium in 2011, and he will be a general chair of CAIP 2017. He is a senior member of the IEEE.View more

Contact IEEE to Subscribe

References

References is not available for this document.