de.mpg.escidoc.pubman.appbase.FacesBean
English
 
Help Guide Disclaimer Contact us Login
  Advanced SearchBrowse

Item

ITEM ACTIONSEXPORT

Released

Conference Paper

How Good are Detection Proposals, really?

MPS-Authors
http://pubman.mpdl.mpg.de/cone/persons/resource/persons86681

Hosang,  Jan
Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons79212

Benenson,  Rodrigo
Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society;

http://pubman.mpdl.mpg.de/cone/persons/resource/persons45383

Schiele,  Bernt
Computer Vision and Multimodal Computing, MPI for Informatics, Max Planck Society;

Locator
There are no locators available
Fulltext (public)

paper082.pdf
(Any fulltext), 2MB

Supplementary Material (public)

sup082.zip
(Supplementary material), 3MB

Citation

Hosang, J., Benenson, R., & Schiele, B. (2014). How Good are Detection Proposals, really? In M. Valstar, A. French, & T. Pridmore (Eds.), Proceedings of the British Machine Vision Conference (pp. 1-12). Durham: BMVA Press.


Cite as: http://hdl.handle.net/11858/00-001M-0000-0024-3C2E-2
Abstract
Current top performing Pascal VOC object detectors employ detection proposals to guide the search for objects thereby avoiding exhaustive sliding window search across images. Despite the popularity of detection proposals, it is unclear which trade‐offs are made when using them during object detection. We provide an in depth analysis of ten object proposal methods along with four baselines regarding ground truth annotation recall (on Pascal VOC 2007 and ImageNet 2013), repeatability, and impact on DPM detector performance. Our findings show common weaknesses of existing methods, and provide insights to choose the most adequate method for different settings.