Tampere University of Technology

TUTCRIS Research Portal

CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

CDTB : A Color and Depth Visual Object Tracking Dataset and Benchmark. / Lukezic, Alan; Kart, Ugur; Käpylä, Jani; Durmush, Ahmed; Kämäräinen, Joni-Kristian; Matas, Jiri; Kristan, Matej.

2019 International Conference on Computer Vision, ICCV 2019. IEEE, 2019. p. 10012-10021 9010284 (IEEE International Conference on Computer Vision).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Lukezic, A, Kart, U, Käpylä, J, Durmush, A, Kämäräinen, J-K, Matas, J & Kristan, M 2019, CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark. in 2019 International Conference on Computer Vision, ICCV 2019., 9010284, IEEE International Conference on Computer Vision, IEEE, pp. 10012-10021, IEEE/CVF International Conference on Computer Vision, 27/10/19. https://doi.org/10.1109/ICCV.2019.01011

APA

Lukezic, A., Kart, U., Käpylä, J., Durmush, A., Kämäräinen, J-K., Matas, J., & Kristan, M. (2019). CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark. In 2019 International Conference on Computer Vision, ICCV 2019 (pp. 10012-10021). [9010284] (IEEE International Conference on Computer Vision). IEEE. https://doi.org/10.1109/ICCV.2019.01011

Vancouver

Lukezic A, Kart U, Käpylä J, Durmush A, Kämäräinen J-K, Matas J et al. CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark. In 2019 International Conference on Computer Vision, ICCV 2019. IEEE. 2019. p. 10012-10021. 9010284. (IEEE International Conference on Computer Vision). https://doi.org/10.1109/ICCV.2019.01011

Author

Lukezic, Alan ; Kart, Ugur ; Käpylä, Jani ; Durmush, Ahmed ; Kämäräinen, Joni-Kristian ; Matas, Jiri ; Kristan, Matej. / CDTB : A Color and Depth Visual Object Tracking Dataset and Benchmark. 2019 International Conference on Computer Vision, ICCV 2019. IEEE, 2019. pp. 10012-10021 (IEEE International Conference on Computer Vision).

Bibtex - Download

@inproceedings{82ef813eea294386a96c332e945fdc38,
title = "CDTB: A Color and Depth Visual Object Tracking Dataset and Benchmark",
abstract = "We propose a new color-and-depth general visual object tracking benchmark (CDTB). CDTB is recorded by several passive and active RGB-D setups and contains indoor as well as outdoor sequences acquired in direct sunlight. The CDTB dataset is the largest and most diverse dataset in RGB-D tracking, with an order of magnitude larger number of frames than related datasets. The sequences have been carefully recorded to contain significant object pose change, clutter, occlusion, and periods of long-term target absence to enable tracker evaluation under realistic conditions. Sequences are per-frame annotated with 13 visual attributes for detailed analysis. Experiments with RGB and RGB-D trackers show that CDTB is more challenging than previous datasets. State-of-the-art RGB trackers outperform the recent RGB-D trackers, indicating a large gap between the two fields, which has not been previously detected by the prior benchmarks. Based on the results of the analysis we point out opportunities for future research in RGB-D tracker design.",
author = "Alan Lukezic and Ugur Kart and Jani K{\"a}pyl{\"a} and Ahmed Durmush and Joni-Kristian K{\"a}m{\"a}r{\"a}inen and Jiri Matas and Matej Kristan",
note = "INT=comp,{"}K{\"a}pyl{\"a}, Jani{"} EXT={"}Matas, Jiri{"} jufoid=58047",
year = "2019",
doi = "10.1109/ICCV.2019.01011",
language = "English",
isbn = "978-1-7281-4804-5",
series = "IEEE International Conference on Computer Vision",
publisher = "IEEE",
pages = "10012--10021",
booktitle = "2019 International Conference on Computer Vision, ICCV 2019",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - CDTB

T2 - A Color and Depth Visual Object Tracking Dataset and Benchmark

AU - Lukezic, Alan

AU - Kart, Ugur

AU - Käpylä, Jani

AU - Durmush, Ahmed

AU - Kämäräinen, Joni-Kristian

AU - Matas, Jiri

AU - Kristan, Matej

N1 - INT=comp,"Käpylä, Jani" EXT="Matas, Jiri" jufoid=58047

PY - 2019

Y1 - 2019

N2 - We propose a new color-and-depth general visual object tracking benchmark (CDTB). CDTB is recorded by several passive and active RGB-D setups and contains indoor as well as outdoor sequences acquired in direct sunlight. The CDTB dataset is the largest and most diverse dataset in RGB-D tracking, with an order of magnitude larger number of frames than related datasets. The sequences have been carefully recorded to contain significant object pose change, clutter, occlusion, and periods of long-term target absence to enable tracker evaluation under realistic conditions. Sequences are per-frame annotated with 13 visual attributes for detailed analysis. Experiments with RGB and RGB-D trackers show that CDTB is more challenging than previous datasets. State-of-the-art RGB trackers outperform the recent RGB-D trackers, indicating a large gap between the two fields, which has not been previously detected by the prior benchmarks. Based on the results of the analysis we point out opportunities for future research in RGB-D tracker design.

AB - We propose a new color-and-depth general visual object tracking benchmark (CDTB). CDTB is recorded by several passive and active RGB-D setups and contains indoor as well as outdoor sequences acquired in direct sunlight. The CDTB dataset is the largest and most diverse dataset in RGB-D tracking, with an order of magnitude larger number of frames than related datasets. The sequences have been carefully recorded to contain significant object pose change, clutter, occlusion, and periods of long-term target absence to enable tracker evaluation under realistic conditions. Sequences are per-frame annotated with 13 visual attributes for detailed analysis. Experiments with RGB and RGB-D trackers show that CDTB is more challenging than previous datasets. State-of-the-art RGB trackers outperform the recent RGB-D trackers, indicating a large gap between the two fields, which has not been previously detected by the prior benchmarks. Based on the results of the analysis we point out opportunities for future research in RGB-D tracker design.

UR - http://openaccess.thecvf.com/content_ICCV_2019/papers/Lukezic_CDTB_A_Color_and_Depth_Visual_Object_Tracking_Dataset_and_ICCV_2019_paper.pdf

U2 - 10.1109/ICCV.2019.01011

DO - 10.1109/ICCV.2019.01011

M3 - Conference contribution

SN - 978-1-7281-4804-5

T3 - IEEE International Conference on Computer Vision

SP - 10012

EP - 10021

BT - 2019 International Conference on Computer Vision, ICCV 2019

PB - IEEE

ER -