Abstract
Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods
and the “perfect single frame detector”. We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech
pedestrian dataset). After manually clustering the frequent errors of a top detector, we characterise both localisation and background-
versus-foreground errors.
To address localisation errors we study the impact of training annotation noise on the detector performance, and show that we can
improve results even with a small portion of sanitised training data. To address background/foreground discrimination, we study convnets
for pedestrian detection, and discuss which factors affect their performance.
Other than our in-depth analysis, we report top performance on the Caltech pedestrian dataset, and provide a new sanitised set of
training and test annotations.
BibTeX
@article{ZBOHS2017, TITLE = {Towards Reaching Human Performance in Pedestrian Detection}, AUTHOR = {Zhang, Shanshan and Benenson, Rodrigo and Omran, Mohamed and Hosang, Jan and Schiele, Bernt}, LANGUAGE = {eng}, ISSN = {0162-8828}, DOI = {10.1109/TPAMI.2017.2700460}, PUBLISHER = {IEEE Computer Society}, ADDRESS = {Los Alamitos, CA}, YEAR = {2018}, DATE = {2018}, ABSTRACT = {Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods and the {\textquotedblleft}perfect single frame detector{\textquotedblright}. We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech pedestrian dataset). After manually clustering the frequent errors of a top detector, we characterise both localisation and background- versus-foreground errors. To address localisation errors we study the impact of training annotation noise on the detector performance, and show that we can improve results even with a small portion of sanitised training data. To address background/foreground discrimination, we study convnets for pedestrian detection, and discuss which factors affect their performance. Other than our in-depth analysis, we report top performance on the Caltech pedestrian dataset, and provide a new sanitised set of training and test annotations.}, JOURNAL = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, VOLUME = {40}, NUMBER = {4}, PAGES = {973--986}, }
Endnote
%0 Journal Article %A Zhang, Shanshan %A Benenson, Rodrigo %A Omran, Mohamed %A Hosang, Jan %A Schiele, Bernt %+ Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society %T Towards Reaching Human Performance in Pedestrian Detection : %G eng %U http://hdl.handle.net/11858/00-001M-0000-002D-440B-2 %R 10.1109/TPAMI.2017.2700460 %7 2017-05-02 %D 2018 %X Encouraged by the recent progress in pedestrian detection, we investigate the gap between current state-of-the-art methods and the “perfect single frame detector”. We enable our analysis by creating a human baseline for pedestrian detection (over the Caltech pedestrian dataset). After manually clustering the frequent errors of a top detector, we characterise both localisation and background- versus-foreground errors. To address localisation errors we study the impact of training annotation noise on the detector performance, and show that we can improve results even with a small portion of sanitised training data. To address background/foreground discrimination, we study convnets for pedestrian detection, and discuss which factors affect their performance. Other than our in-depth analysis, we report top performance on the Caltech pedestrian dataset, and provide a new sanitised set of training and test annotations. %J IEEE Transactions on Pattern Analysis and Machine Intelligence %O IEEE Trans. Pattern Anal. Mach. Intell. TPAMI %V 40 %N 4 %& 973 %P 973 - 986 %I IEEE Computer Society %C Los Alamitos, CA %@ false