Scipediacontent: Scipediacontent moved page Draft Content 363172573 to Eggert et al 2017a

2021-02-08T11:25:53Z

Scipediacontent moved page Draft Content 363172573 to Eggert et al 2017a

Scipediacontent: Created page with " == Abstract == Faster R-CNN is a well-known approach for object detection which combines the generation of region proposals and their classification into a single pipeline...."

2021-02-08T11:25:50Z

Created page with " == Abstract == Faster R-CNN is a well-known approach for object detection which combines the generation of region proposals and their classification into a single pipeline...."

New page

== Abstract ==

Faster R-CNN is a well-known approach for object detection which combines the generation of region proposals and their classification into a single pipeline. In this paper we apply Faster R-CNN to the task of company logo detection. Motivated by the weak performance of Faster R-CNN on small object instances, we perform a detailed examination of both the proposal and the classification stage, examining their behavior for a wide range of object sizes. Additionally, we look at the influence of feature map resolution on the performance of those stages. We introduce an improved scheme for generating anchor proposals and propose a modification to Faster R-CNN which leverages higher-resolution feature maps for small objects. We evaluate our approach on the Flicker data set improving the detection performance on small object instances.

== Original document ==

The different versions of the original document can be found in:

* [http://dx.doi.org/10.1109/icme.2017.8019550 http://dx.doi.org/10.1109/icme.2017.8019550]

* [http://xplorestaging.ieee.org/ielx7/8014303/8019290/08019550.pdf?arnumber=8019550 http://xplorestaging.ieee.org/ielx7/8014303/8019290/08019550.pdf?arnumber=8019550],
: [http://dx.doi.org/10.1109/icme.2017.8019550 http://dx.doi.org/10.1109/icme.2017.8019550]

* [https://ieeexplore.ieee.org/abstract/document/8019550 https://ieeexplore.ieee.org/abstract/document/8019550],
: [https://dblp.uni-trier.de/db/conf/icmcs/icme2017.html#EggertBWZL17 https://dblp.uni-trier.de/db/conf/icmcs/icme2017.html#EggertBWZL17],
: [http://ieeexplore.ieee.org/document/8019550 http://ieeexplore.ieee.org/document/8019550],
: [https://doi.org/10.1109/ICME.2017.8019550 https://doi.org/10.1109/ICME.2017.8019550],
: [https://academic.microsoft.com/#/detail/2752825721 https://academic.microsoft.com/#/detail/2752825721]

* [https://opus.bibliothek.uni-augsburg.de/opus4/frontdoor/index/index/docId/61056 https://opus.bibliothek.uni-augsburg.de/opus4/frontdoor/index/index/docId/61056],
: [https://doi.org/10.1109/icme.2017.8019550 https://doi.org/10.1109/icme.2017.8019550]

← Older revision	Revision as of 11:25, 8 February 2021
(No difference)

Eggert et al 2017a - Revision history

Scipediacontent: Scipediacontent moved page Draft Content 363172573 to Eggert et al 2017a

Scipediacontent: Created page with " == Abstract == Faster R-CNN is a well-known approach for object detection which combines the generation of region proposals and their classification into a single pipeline...."