Master Degree / Yüksek Lisans Tezleri

Permanent URI for this collectionhttps://hdl.handle.net/11147/3008

Browse

Filters

2017 - 201912020 - 20221

Settings

Scopus Q: search.filters.scopusQuartile.N/ASubject: search.filters.subject.Convolutional neural networks

Search Results

Now showing 1 - 2 of 2
  • Master Thesis
    Hierarchical Image Classification With Self-Supervised Vision Transformer Features
    (Izmir Institute of Technology, 2022) Karagüler, Caner; Özuysal, Mustafa
    There are lots of works about image classification and most of them are based on convolutional neural networks (CNN). In image classification, some classes are more difficult to distinguish than others because of non-even visual separability. These difficult classes require domain-specific classifiers but traditional convolutional neural networks are trained as flat N-way classifiers. These flat classifiers can not leverage the hierarchical information of the classes well. To solve this issue, researchers proposed new techniques that embeds class-hierarchy into the convolutional neural networks and most of these techniques exceed existing convolutional neural networks' success rates on large-scale datasets like ImageNet. In this work, we questioned if a hierarchical image classification with self- supervised vision transformer features can exceed hierarchical convolutional neural networks. During this work, we used a hierarchical ETHEC dataset and extract attention features with the help of vision transformers. Using these attention features, we implemented 3 different hierarchical classification approaches and compared the results with CNN alternative of our approaches.
  • Master Thesis
    Localization of Certain Animal Species in Images Via Training Neural Networks With Image Patches
    (Izmir Institute of Technology, 2017) Orhan, Semih; Baştanlar, Yalın
    Object detection is one of the most important tasks for computer vision systems. Varying object size, varying view angle, illumination conditions, occlusion etc. effect the success rate. In recent years, convolutional neural networks (CNNs) have shown great performance in different problems of computer vision including object detection and localization. In this work, we propose a novel training approach for CNNs to localize some animal species whose bodies have distinctive pattern, such as speckles of leopards, black-white lines of zebras, etc. To learn characteristic patterns, small patches are taken from different body parts of animals and they are used to train models. To find object location, in a test image, all locations are visited in a sliding window fashion. Crops are fed to CNN, then classification scores of all patches are recorded. To illustrate object location, heat map is generated by the classification scores of the patches. Afterwards, heat maps are converted to binary images and end up with bounding box estimates of objects. The localization performance of our Patch-based training is compared with Faster R-CNN – a state-of-the-art CNN-based object detection and localization algorithm. While evaluating the performances, in addition to the standard precision-recall metric, we use area-precision and area-recall which represent the potential of Patch-based Model better. Experiment results show that the proposed training method has better performance than Faster R-CNN for most of the evaluated classes. We also showed that Patch-based Model can be used with Faster R-CNN to increase its localization performance.