Filters








16 Hits in 1.3 sec

Relative Attribute Classification with Deep Rank SVM [article]

Sara Atito Ali Ahmed, Berrin Yanikoglu
2020 arXiv   pre-print
Relative attributes indicate the strength of a particular attribute between image pairs. We introduce a deep Siamese network with rank SVM loss function, called Deep Rank SVM (DRSVM), in order to decide which one of a pair of images has a stronger presence of a specific attribute. The network is trained in an end-to-end fashion to jointly learn the visual features and the ranking function. We demonstrate the effectiveness of our approach against the state-of-the-art methods on four image
more » ... rk datasets: LFW-10, PubFig, UTZap50K-lexi and UTZap50K-2 datasets. DRSVM surpasses state-of-art in terms of the average accuracy across attributes, on three of the four image benchmark datasets.
arXiv:2009.07717v1 fatcat:sxoedmvn3betxd5ovbyvzr3ita

Deep Convolutional Neural Network Ensembles Using ECOC

Sara Atito Ali Ahmed, Cemre Zor, Muhammad Awais, Berrin Yanikoglu, Josef Kittler
2021 IEEE Access  
SARA ATITO ALI AHMED (Member, IEEE) received the B.Sc. degree in computer science from Ain Shams University, Egypt, in 2011, and the joint M.Sc. degree from Nile University, Egypt, and TU Berlin, Germany  ... 
doi:10.1109/access.2021.3088717 fatcat:yk6uitl6o5f77iutveserzmle4

Multi-Label Networks for Face Attributes Classification

Sara Atito Aly, Berrin Yanikoglu
2018 2018 IEEE International Conference on Multimedia & Expo Workshops (ICMEW)  
Face attributes classification is drawing attention as a research topic with applications in multiple domains, such as video surveillance and social media analysis. In most attribute classification systems in literature, independent classifiers are trained separately for each attribute. In this work, we propose to train attributes in groups based on their localization (head, eyes, nose, cheek, mouth, shoulder, and general areas) in a multi-task learning scenario to speed up the training process
more » ... and to prevent overfitting. We have evaluated the idea of using the location knowledge for a particular attribute group to speed up the network training. Attention is drawn to the area of interest by blurring training images outside the region of interest, fine-tuning the system and freezing the earlier layers before continuing training with original images. Several data augmentation techniques are also performed to reduce overfitting. Our approach outperforms the state-of-the-art of the attributes on the public LFWA dataset, with an average improvement of almost 0.7% points. The accuracy ranges from 78% (detecting oval face or shadow on the face) to 97.4% (detecting blond hair) across the attributes.
doi:10.1109/icmew.2018.8551518 dblp:conf/icmcs/AlyY18 fatcat:67vxj3m2vrforb5tamu55fub3e

Deep Convolutional Neural Network Ensembles using ECOC [article]

Sara Atito Ali Ahmed, Cemre Zor, Berrin Yanikoglu, Muhammad Awais, Josef Kittler
2021 arXiv   pre-print
Deep neural networks have enhanced the performance of decision making systems in many applications including image understanding, and further gains can be achieved by constructing ensembles. However, designing an ensemble of deep networks is often not very beneficial since the time needed to train the networks is very high or the performance gain obtained is not very significant. In this paper, we analyse error correcting output coding (ECOC) framework to be used as an ensemble technique for
more » ... p networks and propose different design strategies to address the accuracy-complexity trade-off. We carry out an extensive comparative study between the introduced ECOC designs and the state-of-the-art ensemble techniques such as ensemble averaging and gradient boosting decision trees. Furthermore, we propose a combinatory technique which is shown to achieve the highest classification performance amongst all.
arXiv:2009.02961v2 fatcat:6avvpsnpgrbtblnboxbuhkjy5u

Skin Lesion Classification With Deep CNN Ensembles

Sara Atito Ali Ahmed, Berrin Yanikoglu, Ozgu Goksu, Erchan Aptoula
2020 2020 28th Signal Processing and Communications Applications Conference (SIU)  
Early detection of skin cancer is vital when treatment is most likely to be successful. However, diagnosis of skin lesions is a very challenging task due to the similarities between lesions in terms of appearance, location, color, and size. We present a deep learning method for skin lesion classification by fusing and fine-tuning three pre-trained deep learning architectures (Xception, Inception-ResNet-V2, and NasNetLarge) using training images provided by ISIC2019 organizers. Additionally, the
more » ... outliers and the heavy class imbalance are addressed to further enhance the classification of the lesion. The experimental results show that the proposed framework obtained promising results that are comparable with the ISIC2019 challenge leader board.
doi:10.1109/siu49456.2020.9302125 fatcat:3jdhrti5njb25b3yfazeidwsyu

COVID-19 Detection in Computed Tomography Images with 2D and 3D Approaches [article]

Sara Atito Ali Ahmed and Mehmet Can Yavuz and Mehmet Umut Sen and Fatih Gulsen and Onur Tutar and Bora Korkmazer and Cesur Samanci and Sabri Sirolu and Rauf Hamid and Ali Ergun Eryurekli and Toghrul Mammadov and Berrin Yanikoglu
2021 arXiv   pre-print
Detecting COVID-19 in computed tomography (CT) or radiography images has been proposed as a supplement to the definitive RT-PCR test. We present a deep learning ensemble for detecting COVID-19 infection, combining slice-based (2D) and volume-based (3D) approaches. The 2D system detects the infection on each CT slice independently, combining them to obtain the patient-level decision via different methods (averaging and long-short term memory networks). The 3D system takes the whole CT volume to
more » ... rrive to the patient-level decision in one step. A new high resolution chest CT scan dataset, called the IST-C dataset, is also collected in this work. The proposed ensemble, called IST-CovNet, obtains 90.80% accuracy and 0.95 AUC score overall on the IST-C dataset in detecting COVID-19 among normal controls and other types of lung pathologies; and 93.69% accuracy and 0.99 AUC score on the publicly available MosMed dataset that consists of COVID-19 scans and normal controls only. The system is deployed at Istanbul University Cerrahpasa School of Medicine.
arXiv:2105.08506v2 fatcat:phcgpojmgje6hn24xbpdd2nojy

COVID-19 Detection in Computed Tomography Images with 2D and 3D Approaches [article]

Sara Atito Ali Ahmed, Mehmet Can Yavuz, Mehmet Umut Sen, Fatih Gulsen, Onur Tutar, Bora Korkmazer, Cesur Samanci, Sabri Sirolu, Rauf Hamid, Ali Ergun Eryurekli, Toghrul Mammadov, Berrin Yanikoglu
2021
Detecting COVID-19 in computed tomography (CT) or radiography images has been proposed as a supplement to the definitive RT-PCR test. We present a deep learning ensemble for detecting COVID-19 infection, combining slice-based (2D) and volume-based (3D) approaches. The 2D system detects the infection on each CT slice independently, combining them to obtain the patient-level decision via different methods (averaging and long-short term memory networks). The 3D system takes the whole CT volume to
more » ... rrive to the patient-level decision in one step. A new high resolution chest CT scan dataset, called the IST-C dataset, is also collected in this work. The proposed ensemble, called IST-CovNet, obtains 90.80% accuracy and 0.95 AUC score overall on the IST-C dataset in detecting COVID-19 among normal controls and other types of lung pathologies; and 93.69% accuracy and 0.99 AUC score on the publicly available MosMed dataset that consists of COVID-19 scans and normal controls only. The system is deployed at Istanbul University Cerrahpasa School of Medicine.
doi:10.48550/arxiv.2105.08506 fatcat:6jtdcn3qdzhqfaeu4h2r3ncb5i

Author List

2021 2021 29th Signal Processing and Communications Applications Conference (SIU)  
Atito Ali Ahmed YFCC1M Yüz Özellikleri Veri Kümesi saraatito@sabanciuniv.edu 1102 Seda Kul Görme Engelli Bireyler için TensorFlow Lite Kullanılarak Geliştirilen Android İşletim Sistemi Tabanlı Mobil  ...  Veri Kümesi mefe@sabanciuniv.edu 1101 Berrin Yanikoglu YFCC1M Yüz Özellikleri Veri Kümesi berrin@sabanciuniv.edu 1101 Hasan Ocak YFCC1M Yüz Özellikleri Veri Kümesi ocakhasan@sabanciuniv.edu 1101 Sara  ... 
doi:10.1109/siu53274.2021.9477935 fatcat:pxxmgn2cizgmflk3ctifssgzya

Author List

2020 2020 28th Signal Processing and Communications Applications Conference (SIU)  
Emre Pusane Evrisimsel Sinir Aglari ile Veri-Desteksiz Gözü Kapali Isaret Gürültü Orani Kestirimi ali.pusane@boun.edu.tr 1381 Sara Atito Ali Ahmed Skin Lesion Classification With Deep CNN Ensembles  ...  Konusma Varliginin Saptanmasi rasim.ulgen@sestek.com 1581 Murat SARAÇLAR Yapay Sinir Aglari ve Konusmaci Bölütleme ile Gürültülü Ortamlarda Konusma Varliginin Saptanmasi murat.saraclar@boun.edu.tr 1583 Ali  ... 
doi:10.1109/siu49456.2020.9302340 fatcat:bzwjnqtcwjf3tcgzauhq32mx4u

La natura della sentenza canonica di nullità di matrimonio

Enrico Paleari
2017
Ma di ciò sarà espressamente trattato a suo luogo.  ...  Esse, a quanto indirettamente risulta, verrebbero superate dalla opinione corrente col dire che il matrimonio dichiarato inesistente, in realtà non è mai esi• atito, salvi quegli effetti dei quali è appunto  ... 
doi:10.14276/1825-1676.650 fatcat:royl53dnojgzliz3esw7kbml6e

O problema da letra de Vespúcio resolvido pelo encôntro e identificação de 14 dos seus escritos

Roberto Levillier
1955 Revista de História  
E sara una carta in figura piana e un apamundo in corpo sferico".  ...  Lorenzo deite paro- le dj Sua Maestà stimo sarà piú partichularmente raghuagliato.  ...  Io ho ri-, sposto havere scripto alie V.  ... 
doi:10.11606/issn.2316-9141.v11i24p444-498 fatcat:rl5o4jnxrfh5pcul5eyttlninq

Contribuição à Geologia da Bacia de São Paulo

Viktor Leinz, Anna Maria Vieira de Carvalho
1957 Boletim da Faculdade de Filosofia Ciências e Letras Universidade de São Paulo Geologia  
Para tal, utilizamos o mapa topográfico do Município de São Paulo, executado pela Emprêsa Sara Brasil S.  ...  Infelizmente nenhu ma das sondagens localizadas nessa área da depressão atingiram o E m basamento Cristalino, e assim sendo, não foi possível examinarmos o ma terial que ali se encontra recoberto por pacote  ... 
doi:10.11606/issn.2526-3862.bffcluspgeologia.1957.121841 fatcat:hzgmyspbozanlm46glhs3ktudq

Az Orvosi Hetilap 1961 decemberi lapszámai

1961 Orvosi Hetilap  
A g. gl. ezek szerint védte e z t a 2 gyerm eket a serum hep atito g en sajátsága ellen.  ...  Az optim ális zóna általában a 4-7. csövek között jelentkezett.  ...  Tapasztalatunk sze rin t gyakori kis étk e zésekkel és gastropin eg y ü ttes adásával a gastroin testin ális panaszok lényegesen m érsékelhetők.  ... 
doi:10.1556/650.1961.12.03 fatcat:mptqavls4nh2ves3qllhcg6eim

SOCIETÀ LIGURE ATTI D E L L A SOCIETÀ LIGURE

D Storia, Patria Lu M E X V, I Genova, Di, Prayer, Di Vostre, Signorie Serenissime
2011 unpublished
Et d' ali' hora in qua sono stati elletti .  ...  T ali proteste non sono credute sincere da coloro che tacciano il Cromwell d'ipocrita.  ... 
fatcat:s7znjyo7erahflo2g6pd4ua474

GIORNALE STORICO E LETTERARIO DELLA LIGURIA D i r e t t o r e : ARTURO CODIGNOLA MERCANTI MILANESI A GENOVA NEL SECOLO XII Società Ligure di Storia Patria-biblioteca digitale-2012 GIANLUIGI BARNI

Comitato Di, Carlo Bornate -Pietro, Nurra -Vito Vitale
unpublished
In me voi non potete amare che le idee in nome d elle q u a li ho per oltre trent'anni combattuto, com'io potei, e p atito e sperato .  ...  diversi studiosi, il compito degli storici futuri sarà molto facilitato .  ...  fondamentale, utilissim a a lla conoscenza dei personaggi, spes so ignorati oggi da chi non voglia risalire a lla fonte ricchissima, ma un po' faticosa, dei critici nostri del passato : Sopranis, R atti, Ali  ... 
fatcat:feinpxt7rzbshemzmwoalw2mhu
« Previous Showing results 1 — 15 out of 16 results