Veysel Yucesoy

MARVEL: A Large-Scale Image Dataset for Maritime Vessels

Fine-grained visual categorization has recently received great attention as the volumes of the labelled datasets for classification of specific objects, such as cars, bird species, and aircrafts, have been increasing. The collection of large datasets has helped vision based classification approaches and led to significant improvements in performances of the state-of-the-art methods. Visual classification of maritime vessels is another important task assisting naval security and surveillance applications. In this work, we introduce a large-scale image dataset for maritime vessels, consisting of 2 million user uploaded images and their attributes including vessel identity, type, photograph category and year of built, collected from a community website. We categorize the images into 109 vessel type classes and construct 26 superclasses by combining heavily populated classes with a semi-automatic clustering scheme. For the analysis of our dataset, extensive experiments have been performed, involving four potentially useful applications; vessel classification, verification, retrieval, and recognition. We report encouraging results for each application. The introduced dataset is publicly available.

Fine-grained recognition of maritime vessels and land vehicles by deep feature embedding

Recent advances in large-scale image and video analysis have empowered the potential capabilities of visual surveillance systems. In particular, deep learning-based approaches bring in substantial benefits in solving certain computer vision problems such as fine-grained object recognition. Here, the authors mainly concentrate on classification and identification of maritime vessels and land vehicles, which are the key constituents of visual surveillance systems. Employing publicly available data sets for maritime vessels and land vehicles, the authors aim to improve visual recognition. Specifically, the authors focus on five tasks regarding visual recognition; coarse-grained classification, fine-grained classification, coarse-grained retrieval, fine-grained retrieval, and verification. To increase the performance in these tasks, the authors utilise a multi-task learning framework and present a novel loss function which simultaneously considers deep feature learning and classification by exploiting the available hierarchical labels of individual samples and the global statistics of distances between the data pairs. The authors observe that the proposed multi-task learning model improves the fine-grained recognition performance on MARVEL and Stanford Cars data sets, compared to training of a model targeting a single recognition task.

Measuring cross-lingual semantic similarity across European languages

This paper studies cross-lingual semantic similarity (CLSS) between five European languages (i.e. English, French, German, Spanish and Italian) via unsupervised word embeddings from a cross-lingual lexicon. The vocabulary in each language is projected onto a separate high-dimensional vector space, and these vector spaces are then compared using several different distance measures (i.e., correlation, cosine etc.) to measure their pairwise semantic similarities between these languages. A substantial degree of similarity is observed between the vector spaces learned from corpora of the European languages. Null hypothesis testing and bootstrap methods (by resampling without replacement) are utilized to verify the results.

Generic and attribute-specific deep representations for maritime vessels

Fine-grained visual categorization has recently received great attention as the volumes of labeled datasets for classification of specific objects, such as cars, bird species, and air-crafts, have been increasing. The availability of large datasets led to significant performance improvements in several vision-based classification tasks. Visual classification of maritime vessels is another important task, assisting naval security and surveillance applications. We introduced, MARVEL, a large-scale image dataset for maritime vessels, consisting of 2 million user-uploaded images and their various attributes, including vessel identity, type, category, year built, length, and tonnage, collected from a community website. The images were categorized into vessel type classes and also into superclasses defined by combining semantically similar classes, following a semi-automatic clustering scheme. For the analysis of the presented dataset, extensive experiments have been performed, involving several potentially useful applications: vessel type classification, identity verification, retrieval, and identity recognition with and without prior vessel type knowledge. Furthermore, we attempted interesting problems of visual marine surveillance such as predicting and classifying maritime vessel attributes such as length, summer deadweight, draught, and gross tonnage by solely interpreting the visual content in the wild, where no additional cues such as scale, orientation, or location are provided. By utilizing generic and attribute-specific deep representations for maritime vessels, we obtained promising results for the aforementioned applications.

Fine-grained visual marine vessel classification for coastal surveillance and defense applications

The need for capabilities of automated visual content analysis has substantially increased due to presence of large number of images captured by surveillance cameras. With a focus on development of practical methods for extracting effective visual data representations, deep neural network based representations have received great attention due to their success in visual categorization of generic images. For fine-grained image categorization, a closely related yet a more challenging research problem compared to generic image categorization due to high visual similarities within subgroups, diverse applications were developed such as classifying images of vehicles, birds, food and plants. Here, we propose the use of deep neural network based representations for categorizing and identifying marine vessels for defense and security applications. First, we gather a large number of marine vessel images via online sources grouping them into four coarse categories; naval, civil, commercial and service vessels. Next, we subgroup naval vessels into fine categories such as corvettes, frigates and submarines. For distinguishing images, we extract state-of-the-art deep visual representations and train support-vector-machines. Furthermore, we fine tune deep representations for marine vessel images. Experiments address two scenarios, classification and verification of naval marine vessels. Classification experiment aims coarse categorization, as well as learning models of fine categories. Verification experiment embroils identification of specific naval vessels by revealing if a pair of images belongs to identical marine vessels by the help of learnt deep representations. Obtaining promising performance, we believe these presented capabilities would be essential components of future coastal and on-board surveillance systems.

KELİME TEMSİLLERİ İÇİN TEST PERFORMANSINI GELİŞTİRMEYE YÖNELİK EŞDİZİMLİLİK AĞIRLIKLARININ SEÇİMİ

Bu calisma, matematiksel kelime temsillerinin belirli bir gorev icin performanslarinin en iyilenmesi problemini yeniden ele almaktadir. Sayma tabanli (kelimelerin esdizimlilik istatistiklerini hesaba katan) kelime temsili olusturma yontemlerinde klasik olarak kullanilan sayma agirliklari yerine yenilikci agirliklar onererek analoji ve benzerlik bulma gorevlerinde performans artisi saglamak hedeflenmektedir. Calisma dili olarak Turkce secilmis, derlem olusturulurken Turkce’ye has ek-kok yapilari ek alan her kelime yeni bir kelime gibi kabul edilecek sekilde yorumlanmistir. Onerilen esdizimlilik agirliklarinin performansi degisen parametreye gore analiz edilerek sonuclar calisma icerisinde paylasilmistir.

Deep distance metric learning for maritime vessel identification

This paper addresses the problem of maritime vessel identification by exploiting the state-of-the-art techniques of distance metric learning and deep convolutional neural networks since vessels are the key constituents of marine surveillance. In order to increase the performance of visual vessel identification, we propose a joint learning framework which considers a classification and a distance metric learning cost function. The proposed method utilizes the quadruplet samples from a diverse image dataset to learn the ranking of the distances for hierarchical levels of labeling. The proposed method performs favorably well for vessel identification task against the conventional use of neuron activations towards the final layers of the classification networks. The proposed method achieves 60 percent vessel identification accuracy for 3965 different vessels without sacrificing vessel type classification accuracy.

Automated visual classification of indoor scenes and architectural styles

The ability to automatically categorize a large number of new images that are being uploaded to real estate, furniture, and decoration websites, and personalized search functionality will be a great convenience for the users. In this study, modeling of types and architectural styles of indoor scenes is attempted using visual descriptors of different structures. The performance of the learned models is quantitatively measured on useful applications such as image classification and retrieval.

Semantic similarity between Turkish and European languages using word embeddings

Representation of words coming from vocabulary of a language as real vectors in a high dimensional space is called as word embeddings. Word embeddings are proven to be successful in modelling semantic relations between words and numerous natural language processing applications. Although developed mainly for English, word embeddings perform well for many other languages. In this study, semantic similarity between Turkish (two different corpora) and five basic European languages (English, German, French, Spanish, Italian) is calculated using word embeddings over a fixed vocabulary, obtained results are verified using statistical testing. Also, the effect of using different corpora, and additional preprocess steps on the performance of word embeddings on similarity and analogy test sets prepared for Turkish is studied.

Learning traffic congestion by contextual bandit problems for optimum localization

Optimum localization problem, which has a wide range of application areas in real life such as emergency services, command and control systems, warehouse localization, shipment planning, aims to find the best location to minimize the arrival, response or return time which might be vital in some applications. In most of the cases, uncertainty in traffic is the most challenging issue and in the literature generally it is assumed to obey a priori known stochastic distribution. In this study, problem is defined as the optimum localization of ambulances for emergency services and traffic is modeled to be Markovian to generate context data. Unlike the solution methods in the literature, there exists no mutual information transfer between the model and solution of the problem; thus, a contextual multi-armed bandit learner tries to determine the underlying traffic with simple assumptions. The performance of the bandit algorithm is compared with the performance of a classical estimation method in order to show the effectiveness of the learning approach on the solution of the optimum localization problem.