Výsledky vyhledávání - "Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision"

Infrared Small Target Detection with YOLO: Performance Analysis and Optimization

Autoři: Atrash, Abdulkarim Uğur, Ömür ERTEKİN BOLELLİ, ŞEYDA a další

Přispěvatelé: Atrash, Abdulkarim Uğur, Ömür ERTEKİN BOLELLİ, ŞEYDA a další

Témata: Single-Frame Infrared Small Target Detection, Moving Target Detection, Small Object Detection, SIRST, YOLO, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/115660

Dostupnost: https://hdl.handle.net/11511/115660

Joint Temperature-Luminosity Classification of Stellar Spectra with a Distance-Aware Loss Function ; Mesafeye Duyarlı Kayıp Fonksiyonu ile Yıldız Spektrumlarının Ortak Sıcaklık-Parlaklık Sınıflandırması

Autoři: Sezen, Volga Temizel, Alptekin Department of Data Informatics

Přispěvatelé: Sezen, Volga Temizel, Alptekin Department of Data Informatics

Témata: Stellar Classification, 1D CNN, Morgan-Keenan Grid, Ordinal Classification, Distance Aware Loss, Arbis Subjects::Natural Sciences::Physics::Astronomy and Astrophysics::Stars, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Neural Networks

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/115569

Dostupnost: https://hdl.handle.net/11511/115569

Minimizing Ghosting in High Dynamic Range Images and Videos with Hybrid Approaches and Event Guidance ; Hibrit Yaklaşımlar ve Olay Rehberliği ile Yüksek Dinamik Aralıklı Görüntü ve Videolarda Hayaletlenmeyi En Aza İndirme

Autoři: Alpay, Kadir Cenk Akyüz, Ahmet Oğuz Department of Computer Engineering

Přispěvatelé: Alpay, Kadir Cenk Akyüz, Ahmet Oğuz Department of Computer Engineering

Témata: HDR deghosting, Event cameras, Deep networks, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Graphics::Computer Graphics, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Pattern Recognition and Image Processing

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/116083

Dostupnost: https://hdl.handle.net/11511/116083

Addressing Class Imbalance, Label Uncertainty, and Crispness in Edge Detection ; Kenar Tespitinde Sınıf Dengesizliği, Etiket Belirsizliği ve Kenar Keskinliğinin Ele Alınması

Autoři: Çetinkaya, Bedrettin Akbaş, Emre Kalkan, Sinan a další

Přispěvatelé: Çetinkaya, Bedrettin Akbaş, Emre Kalkan, Sinan a další

Témata: Edge Detection, Crisp Edge Detection, Ranking-based Losses, Label Uncertainty, Hyperbolic Embeddings, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Neural Networks

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/116165

Dostupnost: https://hdl.handle.net/11511/116165

Modeling Disease Progression with Diffusion-Based Generative Models ; Difüzyon Tabanlı Üretken Modellerle Hastalık İlerleyişinin Modellenmesi

Autoři: KURT, Meryem Mine Temizel, Alptekin Department of Modeling and Simulation

Přispěvatelé: KURT, Meryem Mine Temizel, Alptekin Department of Modeling and Simulation

Témata: medical image synthesis, diffusion models, ulcerative colitis, disease progression modeling, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/115556

Dostupnost: https://hdl.handle.net/11511/115556

Semi supervised orientation estimation of honeybees ; Bal arılarının yarı denetimli yönelim tahmini

Autoři: Gündeğer, bilal yağız Şahin, Erol Department of Computer Engineering

Přispěvatelé: Gündeğer, bilal yağız Şahin, Erol Department of Computer Engineering

Témata: Markerless tracking, Orientation estimation, Lifelong learning, Honeybee, Behavior, Semi-supervised learning, Arbis Subjects::Engineering and Technology, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition, Computer Learning and Pattern Recognition::Computer Learning, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/115538

Dostupnost: https://hdl.handle.net/11511/115538

UNSUPERVISED AND SEMI-SUPERVISED DOMAIN ADAPTATION FOR SEMANTIC SEGMENTATION ; ANLAMSAL BÖLÜTLEME İÇİN DENETİMSİZ VE YARI DENETİMLİ ALAN UYARLAMASI

Autoři: Erce, Beyza Ecem Koçyiğit, Altan Department of Data Informatics

Přispěvatelé: Erce, Beyza Ecem Koçyiğit, Altan Department of Data Informatics

Témata: Semantic segmentation, unsupervised domain adaptation, semi-supervised domain adaptation, DeepLabV3+, DANN, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Other

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/114992

Dostupnost: https://hdl.handle.net/11511/114992

A novel scalable global alignment method for 3D reconstruction ; Üç boyutlu yeniden yapılandırma için yeni bir ölçeklenebilir küresel hizalama yöntemi

Autoři: Üstündaş, Şahin Umutcan Şahin, Erol Akbaş, Emre a další

Přispěvatelé: Üstündaş, Şahin Umutcan Şahin, Erol Akbaş, Emre a další

Témata: 3D reconstruction, Structure-from-Motion, Global alignment, Novel view synthesis, 3D computer vision, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/114597

Dostupnost: https://hdl.handle.net/11511/114597

Minimally Supervised Tracking of Animal Colonies with Iteratively Trained Object Detectors ; Yinelemeli Eğitilen Nesne Algılayıcılar İle Hayvan Kolonilerinin Minimal Denetimli Takibi

Autoři: Gödelek, Oğuz Alemdar, Hande Department of Computer Engineering

Přispěvatelé: Gödelek, Oğuz Alemdar, Hande Department of Computer Engineering

Témata: Animal tracking, Computer vision, High-density object detection, Self-training, Knowledge distillation, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Computer Learning, Computer Learning and Pattern Recognition::Neural Networks

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/113490

Dostupnost: https://hdl.handle.net/11511/113490

Bucketing Ranking-Based Losses For Efficient Training of Object Detectors ; Nesne Tespitçilerinin Etkin Eğitimi İçin Sıralama Bazlı Kayıp Fonksiyonlarını Gruplama

Autoři: Yavuz, Feyza Kalkan, Sinan Akbaş, Emre a další

Přispěvatelé: Yavuz, Feyza Kalkan, Sinan Akbaş, Emre a další

Témata: Object Detection, Detection Transformers, Deep Learning, Ranking-based Loss Functions, Loss Functions, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/113453

Dostupnost: https://hdl.handle.net/11511/113453

UTILIZATION OF GAUSSIAN SPLATTING IN VISUAL SLAM ; GÖRSEL E¸S ZAMANLI KONUMLANDIRMA VE HAR˙ITALAMADA GAUSS SIÇRAMASININ KULLANIMI

Autoři: Sarıkamış, Furkan Aykut Alatan, Abdullah Aydın Computational Design and Fabrication Technologies in Department of Architecture

Přispěvatelé: Sarıkamış, Furkan Aykut Alatan, Abdullah Aydın Computational Design and Fabrication Technologies in Department of Architecture

Témata: Gaussian Splatting, Neural Radiance Fields, Visual SLAM, 3D reconstruction, Arbis Subjects::Engineering and Technology, Arbis Subjects::Engineering and Technology::Electrical and Electronics Engineering::Other, Arbis Subjects::Engineering and Technology::Electrical and Electronics Engineering, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/113440

Dostupnost: https://hdl.handle.net/11511/113440

Investigation of the effects of lithium on locomotor activity, circadian rhythm, and learning using computational methods in honey bees ; Lityumun bal arılarında lokomotor aktivite, sirkadyen ritm ve öğrenme üzerine etkilerinin hesaplamalı yöntemler kullanılarak araştırılması

Autoři: Erdem, Babür Gözen, Ayşe Gül Alemdar, Hande a další

Přispěvatelé: Erdem, Babür Gözen, Ayşe Gül Alemdar, Hande a další

Témata: Honey Bee, Lithium, Locomotor Activity, Circadian Rhythm, Learning, Arbis Subjects::Natural Sciences::Life Sciences::Bioinformatics, Arbis Subjects::Natural Sciences::Life Sciences::Neurobiology::Neuroethology, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/112599

Dostupnost: https://hdl.handle.net/11511/112599

On Calibrating Deep Object Detectors ; Derin Nesne Dedektörlerinin Kalibrasyonu

Autoři: Güngör, Muhammed Ertuğrul Kalkan, Sinan Akbaş, Emre a další

Přispěvatelé: Güngör, Muhammed Ertuğrul Kalkan, Sinan Akbaş, Emre a další

Témata: Object Detection, Long-tailed Learning, Class Score Calibration, Logit Adjustment, Arbis Subjects::Engineering and Technology, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Computer Learning and Pattern Recognition::Neural Networks

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/107826

Dostupnost: https://hdl.handle.net/11511/107826

A novel content-based retrieval system for hyperspectral remote sensing imagery ; Hiperspektral uzaktan algılama görüntüleri için yenilikçi bir içerik tabanlı erişim sistemi

Autoři: Ömrüuzun, Fatih Yardımcı Çetin, Yasemin Leloğlu, Uğur Murat a další

Přispěvatelé: Ömrüuzun, Fatih Yardımcı Çetin, Yasemin Leloğlu, Uğur Murat a další

Témata: Remote sensing, Content-based semantic retrieval, Hyperspectral imaging, Feature extraction, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Pattern Recognition and Image Processing, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Other, Arbis Subjects::Engineering and Technology::Computer Sciences::Database and Data Structures::Information Storage and Retrieval, Arbis Subjects::Engineering and Technology::Information Systems, Communication and Control Engineering::Signal Processing::Image and Video Processing

Popis souboru: 127 pages; application/pdf

Relation: https://hdl.handle.net/11511/109799

Dostupnost: https://hdl.handle.net/11511/109799

Deep learning methods for blind super resolution using self-attention transformers and degradation estimations ; Öz-dikkat dönüştürücüler ve bozulma tahminleri kullanarak kör süper çözünürlük için derin öğrenme yöntemleri

Autoři: Vardar, Batuhan Ertekin Bolelli, Şeyda Department of Computer Engineering

Přispěvatelé: Vardar, Batuhan Ertekin Bolelli, Şeyda Department of Computer Engineering

Témata: Blind super-resolution, Contrastive learning, Degradation learning, Vision transformers, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: 62 pages; application/pdf

Relation: https://hdl.handle.net/11511/108517

Dostupnost: https://hdl.handle.net/11511/108517

EFFICIENT PRETRAINING OF VISION TRANSFORMERS: A LAYER-FREEZING APPROACH WITH LOCAL MASKED IMAGE MODELING ; GÖRSEL DÖNÜŞTÜRÜCÜLERİN VERİMLİ ÖN EĞİTİMİ: YEREL MASKELİ GÖRÜNTÜ MODELLEME İLE KATMAN DONDURMA YAKLAŞIMI

Autoři: Topçuoğlu, Utku Mert Akagündüz, Erdem Department of Modeling and Simulation

Přispěvatelé: Topçuoğlu, Utku Mert Akagündüz, Erdem Department of Modeling and Simulation

Témata: Vision Transformers, Self-Supervised Learning, Local Masked Image Modeling, Progres- sive Layer Freezing, Computational Efficiency, Multi-Scale Reconstruction, Adaptive Learning Rate Scheduling, FreezeOut Learning Rate Scheduling, Training Time Reduction, Arbis Subjects::Engineering and Technology, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Information Systems, Communication and Control Engineering, Communication and Control Engineering::Robotics and Mechatronics Systems::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Neural Networks, Computer Learning and Pattern Recognition::Pattern Recognition and Image Processing

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/111029

Dostupnost: https://hdl.handle.net/11511/111029

ENHANCING DNN TEST DATA SELECTION THROUGH UNCERTAINTY-BASED AND DATA DISTRIBUTION-AWARE APPROACHES ; BELİRSİZLİĞE DAYALI VE VERİ DAĞILIMINA DUYARLI YAKLAŞIMLAR YOLUYLA DNN TESTI VERİ SEÇİMİNİN GELİŞTİRİLMESİ

Autoři: Demir, Demet Sürer, Elif Betin Can, Aysu a další

Přispěvatelé: Demir, Demet Sürer, Elif Betin Can, Aysu a další

Témata: Deep Neural Network Testing, Test Data Selection and Prioritization, Data Distribution, Deep Learning Explainability, Deep Learning Uncertainty, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Neural Networks, Computer Learning and Pattern Recognition::Uncertainity Management, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Software::Software Engineering

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/110146

Dostupnost: https://hdl.handle.net/11511/110146

A crowd-aware and self-supervised approach for object detection in wide area motion imagery ; Geniş alan hareketli görüntülerde nesne tespiti için kendi kendini denetleyen ve kalabalığın farkında yöntem

Autoři: Hatipoğlu, Poyraz Umut İyigün, Cem Kalkan, Sinan a další

Přispěvatelé: Hatipoğlu, Poyraz Umut İyigün, Cem Kalkan, Sinan a další

Témata: Moving object detection, Wide-area motion imagery, Convolutional neural networks, Deep learning, Self-supervised learning, Nonlinear Optimization, Arbis Subjects::Engineering and Technology, Arbis Subjects::Engineering and Technology::Computer Sciences, Arbis Subjects::Engineering and Technology::Industrial Engineering, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition, Computer Learning and Pattern Recognition::Neural Networks, Computer Learning and Pattern Recognition::Pattern Recognition and Image Processing, Arbis Subjects::Engineering and Technology::Industrial Engineering::Optimization Theory and Methods, Arbis Subjects::Engineering and Technology::Industrial Engineering::Optimization Theory and Methods::Nonlinear Programming

Popis souboru: 140 pages; application/pdf

Relation: https://hdl.handle.net/11511/104519

Dostupnost: https://hdl.handle.net/11511/104519

The reusability prior in deep learning models ; Derin öğrenme modellerinde yeniden kullanılabilirlik önseli

Autoři: Polat, Aydın Göze Alpaslan, Ferda Nur Department of Computer Engineering

Přispěvatelé: Polat, Aydın Göze Alpaslan, Ferda Nur Department of Computer Engineering

Témata: Deep learning, Parameter sharing, Genetic algorithms, Reusability, Statistical mechanics, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Neural Networks, Computer Learning and Pattern Recognition::Evolutionary Computing, Arbis Subjects::Natural Sciences::Physics::General Physics::Statistical physics, thermodynamic and nonlinear dynamic systems, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/104170

Dostupnost: https://hdl.handle.net/11511/104170

An extensive analysis on oriented object detection ; Yönlü nesne tespiti üzerine detaylı bir analiz

Autoři: Koç, İbrahim Kalkan, Sinan Department of Computer Engineering

Přispěvatelé: Koç, İbrahim Kalkan, Sinan Department of Computer Engineering

Témata: Object Detection, Oriented Object Detection, Scale and Orientation Imbalance, Arbis Subjects::Engineering and Technology::Computer Sciences::Computer Vision::Computer Vision, Arbis Subjects::Engineering and Technology::Computer Sciences::Artificial Intelligence, Computer Learning and Pattern Recognition::Neural Networks

Popis souboru: application/pdf

Relation: https://hdl.handle.net/11511/107717

Dostupnost: https://hdl.handle.net/11511/107717