METU dataset: A big dataset for benchmarking trademark retrieval

Tursun, Osman; Kalkan, Sinan

doi:10.1109/mva.2015.7153243

Cited by 18 publications

(13 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An empirical analysis shows that some image groups are often related to certain named entities (NE) classes when using search engines, as described in Table 1. For training purposes, we used the Scene 13 dataset [8] to train our classifiers for location (LOC), "faces" from Caltech 101 Object Categories [7] to train our person (PER) and logos from METU dataset [16] for organisation ORG object detection. These datasets produces the training data for our set of supervised classifiers (1 for ORG, 1 for PER and 10 for LOC).…”

Section: Conceptual Architecturementioning

confidence: 99%

Named Entity Recognition in Twitter Using Images and Text

Esteves

Peres

Lehmann

et al. 2018

Current Trends in Web Engineering

View full text Add to dashboard Cite

Named Entity Recognition (NER) is an important subtask of information extraction that seeks to locate and recognise named entities. Despite recent achievements, we still face limitations with correctly detecting and classifying entities, prominently in short and noisy text, such as Twitter. An important negative aspect in most of NER approaches is the high dependency on hand-crafted features and domain-specific knowledge, necessary to achieve state-of-the-art results. Thus, devising models to deal with such linguistically complex contexts is still challenging. In this paper, we propose a novel multi-level architecture that does not rely on any specific linguistic resource or encoded rule. Unlike traditional approaches, we use features extracted from images and text to classify named entities. Experimental tests against state-of-the-art NER for Twitter on the Ritter dataset present competitive results (0.59 F-measure), indicating that this approach may lead towards better NER models.

show abstract

Section: Conceptual Architecturementioning

confidence: 99%

Named Entity Recognition in Twitter Using Images and Text

Esteves

Peres

Lehmann

et al. 2018

Current Trends in Web Engineering

View full text Add to dashboard Cite

show abstract

“…METU Trademark Dataset. The METU dataset [26] is the largest public dataset for TR [27]. It includes 589, 098 text-only marks, 19, 387 figure-only marks and 311, 986 figure and text marks.…”

Section: Datasetsmentioning

confidence: 99%

Component-Based Attention for Large-Scale Trademark Retrieval

Tursun

Denman

Sivapalan

et al. 2022

IEEE Trans.Inform.Forensic Secur.

Self Cite

View full text Add to dashboard Cite

The demand for large-scale trademark retrieval (TR) systems has significantly increased to combat the rise in international trademark infringement. Unfortunately, the ranking accuracy of current approaches using either handcrafted or pre-trained deep convolution neural network (DCNN) features is inadequate for large-scale deployments. We show in this paper that the ranking accuracy of TR systems can be significantly improved by incorporating hard and soft attention mechanisms, which direct attention to critical information such as figurative elements and reduce attention given to distracting and uninformative elements such as text and background. Our proposed approach achieves state-of-the-art results on a challenging large-scale trademark dataset.

show abstract

“…For evaluation, we follow the same evaluation protocol described in [2]. In detail, we first return the ranking results for each query by sorting the similarity scores of the gallery images.…”

Section: Dataset and Evaluation Protocolmentioning

confidence: 99%

“…Later, LSTR using content-based image retrieval (CBIR) algorithms have been used thanks to it's efficiency and accuracy. Hand-crafted features based-on shape, color or texture were developed for early CBIR-LSTR systems [2,3]. With the rise of deep learning, off-the-shelf deep features have been applied for LSTR, demonstrating higher accuracy and efficiency compared to traditional hand-crafted features.…”

Section: Introductionmentioning

confidence: 99%

Learning Regional Attention Over Multi-Resolution Deep Convolutional Features For Trademark Retrieval

Tursun

Denman

Sridharan

et al. 2021

2021 IEEE International Conference on Image Processing (ICIP)

Self Cite

View full text Add to dashboard Cite

Large-scale trademark retrieval is an important content-based image retrieval task. A recent study shows that off-theshelf deep features aggregated with Regional-Maximum Activation of Convolutions (R-MAC) achieve state-of-theart results. However, R-MAC suffers in the presence of background clutter/trivial regions and scale variance, and discards important spatial information. We introduce three simple but effective modifications to R-MAC to overcome these drawbacks. First, we propose the use of both sum and max pooling to minimise the loss of spatial information. We also employ domain-specific unsupervised soft-attention to eliminate background clutter and unimportant regions. Finally, we add multi-resolution inputs to enhance the scaleinvariance of R-MAC. We evaluate these three modifications on the million-scale METU dataset. Our results show that all modifications bring non-trivial improvements, and surpass previous state-of-the-art results.

show abstract

METU dataset: A big dataset for benchmarking trademark retrieval

Cited by 18 publications

References 14 publications

Named Entity Recognition in Twitter Using Images and Text

Named Entity Recognition in Twitter Using Images and Text

Component-Based Attention for Large-Scale Trademark Retrieval

Learning Regional Attention Over Multi-Resolution Deep Convolutional Features For Trademark Retrieval

Contact Info

Product

Resources

About