Deep learning for logo detection

DSpace Repository

Show simple item record Paleček, Karel 2019-10-15T15:25:01Z 2019-10-15T15:25:01Z 2019
dc.description.abstract We present a deep learning system for automatic logo detection in real world images. We base our detector on the popular framework of FasterR-CNN and compare its performance to other models such as Mask R-CNN or RetinaNet. We perform a detailed empirical analysis of various design and architecture choices and show how these can have much higher influence than algorithmic tweaks or popular techniques such as data augmentation. We also provide a systematic detection performance comparison of various models on multiple popular datasets including FlickrLogos-32, TopLogo-10 and recently introduced QMUL-OpenLogo benchmark, which allows for a direct comparison between recently proposed extensions. By careful optimization of the training procedure we were able to achieve significant improvements of the state of the art on all mentioned datasets. We apply our observations to build a detector to detect logos of the Red Bull brand in online media and images. cs
dc.language.iso cs cs
dc.publisher Institute of Electrical and Electronics Engineers Inc.
dc.relation.ispartof 2019 42nd International Conference on Telecommunications and Signal Processing, TSP 2019
dc.subject Faster R-CNN cs
dc.subject FlickrLogos-32 cs
dc.subject Logo detection cs
dc.subject Mask R-CNN cs
dc.subject QMUl-OpenLogo cs
dc.subject RetinaNet cs
dc.subject TopLogo-10 cs
dc.title Deep learning for logo detection cs
dc.type Conference Paper
dc.identifier.doi 10.1109/TSP.2019.8769038
local.citation.spage 609
local.citation.epage 612
local.event.sdate 2019-07-01
local.event.edate 2019-07-03
local.event.title 42nd International Conference on Telecommunications and Signal Processing, TSP 2019
local.event.location Radisson Blu Beke HotelSuperior, 43. Terez krt.Budapest; Hungary
local.identifier.publikace 7160
local.article.number 8769038

Files in this item

This item appears in the following Collection(s)

Show simple item record

Search DSpace

Advanced Search


My Account