WebLogo-2M: Scalable Logo Detection by Deep Learning from the Web

Su, Hang; Gong, Shaogang; Zhu, Xiatian

doi:10.1109/iccvw.2017.41

Cited by 51 publications

(28 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The preliminary version of this has been reported in [24]. Compared with the earlier study, there are several key differences introduced: (i) This study presents a more advanced method by introducing a joint co-training and self-learning concept into the scalable logo detection model formulation.…”

Section: Existing Logo Detection Methods Typically Consider a Small Nmentioning

confidence: 99%

Scalable logo detection by self co-learning

Gong

Zhu³

2020

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

Existing logo detection methods usually consider a small number of logo classes, limited images per class and assume fine-gained object bounding box annotations. This limits their scalability to real-world dynamic applications. In this work, we tackle these challenges by exploring a web data learning principle without the need for exhaustive manual labelling. Specifically, we propose a novel incremental learning approach, called Scalable Logo Self-co-Learning (SL 2), capable of automatically selfdiscovering informative training images from noisy web data for progressively improving model capability in a cross-model co-learning manner. Moreover, we introduce a very large (2,190,757 images of 194 logo classes) logo dataset "WebLogo-2M" by designing an automatic data collection and processing method. Extensive comparative evaluations demonstrate the superiority of SL 2 over the state-of-the-art strongly and weakly supervised detection models and contemporary web data learning approaches.

show abstract

Section: Existing Logo Detection Methods Typically Consider a Small Nmentioning

confidence: 99%

Scalable logo detection by self co-learning

Gong

Zhu³

2020

Pattern Recognition

Self Cite

View full text Add to dashboard Cite

show abstract

“…Relating this prior art with our contributions in logo research; none of these approaches explore logo recognition as a few-shot clustering formulation which we feel is a better fit for this problem. As a result, we did not have to perform class imbalance correction (as done by [59]), our approach can handle large number of logo classes and do effective few-shot logo detection by projecting new logos into an embedding space. We used a combination of triplet-loss and proxies [40] to optimize this embedding space and not a simple distance measure.…”

Section: Metric Learningmentioning

confidence: 99%

Scalable Logo Recognition Using Proxies

Fehervari

Appalaraju

2019

2019 IEEE Winter Conference on Applications of Computer Vision (WACV)

View full text Add to dashboard Cite

Logo recognition is the task of identifying and classifying logos. Logo recognition is a challenging problem as there is no clear definition of a logo and there are huge variations of logos, brands and re-training to cover every variation is impractical. In this paper, we formulate logo recognition as a few-shot object detection problem. The two main components in our pipeline are universal logo detector and fewshot logo recognizer. The universal logo detector is a classagnostic deep object detector network which tries to learn the characteristics of what makes a logo. It predicts bounding boxes on likely logo regions. These logo regions are then classified by logo recognizer using nearest neighbor search, trained by triplet loss using proxies. We also annotated a first of its kind product logo dataset containing 2000 logos from 295K images collected from Amazon called PL2K. Our pipeline achieves 97% recall with 0.6 mAP on PL2K test dataset and state-of-the-art 0.565 mAP on the publicly available FlickrLogos-32 test set without fine-tuning.Accurate logo recognition in images can have multiple applications. It can enable better semantic search, better personalized product recommendations, improved contextual ads, IP infringement detection amongst other applications.Logo recognition has many inter-and intra-class variations, retraining with each new variation is unscalable. In This paper is a preprint (IEEE accepted status at IEEE WACV

show abstract

“…Images FlickLogos-27 [18] 27 1080 FlickLogos-32 [30] 32 8240 BelgaLogos [17] 37 10000 LOGO-Net [14] 160 73414 WebLogo-2M [34] 194 1867177 LLD-icon (ours) 486377 486377 LLD-logo (ours) 122920 122920 LLD (ours) 486377+ 609297 Table 1: Logo datasets. Our LLD provides orders of magnitude more logos than the existing public datasets.…”

Section: Logosmentioning

confidence: 99%

“…Therefore, we crawl a highly diverse datasetthe Large Logo Dataset (LLD) -of real logos 'in the wild' from the Internet. As shown in Table 1 our LLD proposes thousands of times more distinct logos than the largest public logo dataset to date, WebLogo-2M [34].…”

Section: Introduction and Related Workmentioning

confidence: 99%

Logo Synthesis and Manipulation with Clustered Generative Adversarial Networks

Sage

Timofte

Agustsson

et al. 2018

2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition

View full text Add to dashboard Cite

Designing a logo for a new brand is a lengthy and tedious back-and-forth process between a designer and a client. In this paper we explore to what extent machine learning can solve the creative task of the designer. For this, we build a dataset -LLD -of 600k+ logos crawled from the world wide web. Training Generative Adversarial Networks (GANs) for logo synthesis on such multi-modal data is not straightforward and results in mode collapse for some state-of-the-art methods. We propose the use of synthetic labels obtained through clustering to disentangle and stabilize GAN training. We are able to generate a high diversity of plausible logos and we demonstrate latent space exploration techniques to ease the logo design task in an interactive manner. Moreover, we validate the proposed clustered GAN training on CIFAR 10, achieving state-of-theart Inception scores when using synthetic labels obtained via clustering the features of an ImageNet classifier. GANs can cope with multi-modal data by means of synthetic labels achieved through clustering, and our results show the creative potential of such techniques for logo synthesis and manipulation. Our dataset and models will be made publicly available at https

show abstract

WebLogo-2M: Scalable Logo Detection by Deep Learning from the Web

Cited by 51 publications

References 26 publications

Scalable logo detection by self co-learning

Scalable logo detection by self co-learning

Scalable Logo Recognition Using Proxies

Logo Synthesis and Manipulation with Clustered Generative Adversarial Networks

Contact Info

Product

Resources

About