Logo detection with extendibility and discrimination

Li, Kuo-Wei; Chen, Shuyuan; Su, Songzhi; Duh, Der-Jyh; Li, Shaozi

doi:10.1007/s11042-013-1449-1

Cited by 28 publications

(20 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Dataset Logo # Object # Image # PA BelgaLogos [20] 37 2,695 1,951 Yes FlickrLogos-27 [21] 27 4,536 810 Yes FlickrLogos-32 [33] 32 3404 2,240 Yes LOGO-NET [16] 160 130,608 73,414 No is extremely costly [16] and unaffordable in most cases, not only in monetary but more critically in timescale terms. In the current literature, most existing studies on logo detection are limited to small scales, in both the number of logo images and logo classes [18,23], largely due to the high costs in constructing large scale logo datasets. It is non-trivial to collect automatically large scale logo training data that covers a large number of different logo classes.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning Logo Detection with Data Expansion by Synthesising Context

Zhu

Gong

2017

2017 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

Logo detection in unconstrained images is challenging, particularly when only very sparse labelled training images are accessible due to high labelling costs. In this work, we describe a model training image synthesising method capable of improving significantly logo detection performance when only a handful of (e.g., 10) labelled training images captured in realistic context are available, avoiding extensive manual labelling costs. Specifically, we design a novel algorithm for generating Synthetic Context Logo (SCL) training images to increase model robustness against unknown background clutters, resulting in superior logo detection performance. For benchmarking model performance, we introduce a new logo detection dataset TopLogo-10 collected from top 10 most popular clothing/wearable brandname logos captured in rich visual context. Extensive comparisons show the advantages of our proposed SCL model over the state-of-the-art alternatives for logo detection using two real-world logo benchmark datasets: FlickrLogo-32 and our new TopLogo-10 1 .

show abstract

Section: Introductionmentioning

confidence: 99%

Deep Learning Logo Detection with Data Expansion by Synthesising Context

Zhu

Gong

2017

2017 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

show abstract

“…Comparative results on the FlickrLogos-32 database. We report the results achieved with the proposed method, by our older method (Boia&Florea (Boia and Florea, 2015)), and by the methods introduced by Li et al (Li et al, 2014), Lu et al (Lu et al, 2014) and respectively Romberg et al (Romberg and Lienhart, 2013). Figure 13.…”

Section: Resultsmentioning

confidence: 99%

“…Also Li et al (Li et al, 2014) used a Support Vector Machine (SVM) to select the HoG described potential windows of interest and further classify them with affine SIFT and nearest neighbor; yet they still report precision and recall instead of detection rate. A summary of existing works may also be followed in table 1.…”

Section: Introductionmentioning

confidence: 99%

Logo localization and recognition in natural images using homographic class graphs

Boia

Florea

et al. 2015

Machine Vision and Applications

View full text Add to dashboard Cite

We propose a method for localization and classification of brand logos in natural images. The system has to overcome multiple challenges such as perspective deformations, warping, variations of the shape and colors, occlusions, background variations. To deal with perspective variation, we rely on homography matching between the SIFT keypoints of logo instances of the same class. To address the changes in color, we construct a weighted graph of logo interconnections that is further analyzed to extract potentially multiple instances of the class. The main instance is build by grouping the keypoints of the graph connected logos onto the central image. The secondary instance is needed for color inverted logos and is obtained by inverting the orientation of the main instance. The constructed logo recognition system is tested on two databases (FlickrLogos-32 and BelgaLogos), outperforming state of the art with more than 10% accuracy.to be published in Machine Vision and Application. For final version please check on Springer

show abstract

“…To improve the process of finding region candidates, Li, Chen, and Su [19] proposed sliding window scanning technique that produces a set of candidate windows of various sizes. For each candidate windows, histograms of oriented gradient (HOGE) feature are extracted and Support Vector Machine (SVM) model is then used to recognize logo in the respective window.…”

Section: Related Work a Logo Detectionmentioning

confidence: 99%

One-Stage Logo Detection Framework using AdaBoost Resnet50 Backbone

Sarwo*¹,

Heryadi²,

Budiharto³

et al. 2019

IJRTE

View full text Add to dashboard Cite

Logo is an important asset as it is designed to express identity or character of the company or organization that owns the logo. The advent of deep learning methods and proliferated of logo images sample dataset in the past decade has made automated logo detection from digital images or video an interesting computer vision problem with wide potential applications. This paper presents a novel one-stage logo detector framework in which the backbone of the proposed logo detector is a deep learning model which is trained supervisedly using gradient descent training algorithm and the target logo classes as input dataset. The experiment results showed that AdaBoost Resnet50 (0.58 MAP) as the logo detector backbone outperforms Resnet50 (0.56 MAP), VGG19 (0.32 MAP), and AdaBoost VGG19 (0.56 MAP).

show abstract

Logo detection with extendibility and discrimination

Cited by 28 publications

References 29 publications

Deep Learning Logo Detection with Data Expansion by Synthesising Context

Deep Learning Logo Detection with Data Expansion by Synthesising Context

Logo localization and recognition in natural images using homographic class graphs

One-Stage Logo Detection Framework using AdaBoost Resnet50 Backbone

Contact Info

Product

Resources

About