Phd Degree / Doktora

Permanent URI for this collectionhttps://hdl.handle.net/11147/2869

Browse

Filters

2015 - 201912020 - 20232

Settings

Subject: search.filters.subject.Computer visionWoS Q: search.filters.wosQuartile.N/A

Search Results

Now showing 1 - 3 of 3
  • Doctoral Thesis
    Classification of Maneuvers of Vehicles in Front for Driver Assistance Systems
    (01. Izmir Institute of Technology, 2023) Nalçakan, Yağız; Baştanlar, Yalın
    Predicting vehicle maneuvers is a critical task for developing autonomous driving. These maneuvers have been identified as leading causes of fatal accidents, underscoring the need for robust and reliable detection systems. This thesis addresses this critical issue by developing and evaluating novel methodologies for classifying maneuvers, especially lane change and cut-in maneuvers in front of the vehicle. Two specific methods are proposed in this thesis work, and their effectiveness is evaluated on two datasets: the Prevention Lane Change Prediction dataset and the BDD-100K Cut-in/Lane-pass Classification Subset. The first method is a model that utilizes features extracted from the bounding boxes of the target vehicle, feeding them into a single-layer LSTM network for cut-in/lane-pass classification. The second method involves training a 3-dimensional residual neural network in a self-supervised manner using contrastive video representation learning. For the self-supervised training phase, a novel scene representation is proposed to highlight vehicle motions. Afterward, the same model is fine-tuned using labeled video data. Lastly, an ensemble learning approach is introduced, which combines the predictive capabilities of the proposed LSTM-based and self-supervised contrastive video representation learning models, leveraging the strengths of both methods to enhance the overall maneuver classification performance. The proposed methods made significant contributions to the field. The LSTM-based model achieved high classification accuracies compared to other studies in the literature. The self-supervised video representation learning model represents the first application of contrastive learning in maneuver classification. The ensemble learning approach has shown a significant improvement in the performance of the maneuver detection system.
  • Doctoral Thesis
    Planar Geometry Estimation With Deep Learning
    (Izmir Institute of Technology, 2022) Uzyıldırım, Furkan Eren; Özuysal, Mustafa
    Understanding the geometric structure of any scene is one of the oldest problems in Computer Vision. Most scenes include planar regions that provide information about the geometric structure and their automatic detection and segmentation plays an important role in many computer vision applications. In recent years, convolutional neural network architectures have been introduced for piece-wise planar segmentation. They outperform the traditional approaches that generate plane candidates with 3D segmentation methods from the explicitly reconstructed 3D point cloud. However, most of the convolutional neural network architectures are not designed and trained for outdoor scenes, because they require manual annotation, which is a time-consuming task that results in a lack of training data. In this thesis,we propose and develop a deep learning based framework for piece-wise plane detection and segmentation of outdoor scenes without requiring manually annotated training data. We exploit a network trained on imagery with annotated targets and an automatically reconstructed point cloud from either Structure from Motion-Multi View Stereo pipeline or monocular depth estimation network to estimate the training ground truth on the outdoor images in an iterative energy minimization framework. We show that the resulting ground truth estimate of various sets of images in the outdoor domain is good enough to improve network weights of different architectures trained on ground truth annotated images. Moreover, we demonstrate that this transfer learning scheme can be repeated multiple times iteratively to further improve the accuracy of plane detection and segmentation on monocular images of outdoor scenes.
  • Doctoral Thesis
    Joint Reconstruction of Surface Geometry and Reflection Properties by Using Image Based Methods
    (Izmir Institute of Technology, 2015) Ozan, Şükrü; Gümüştekin, Şevket
    In this thesis, we aim to capture realistic geometrical descriptions of real world scenes and objects with a special effort to characterize reflection properties. After a brief review of the stereo imaging literature, we show our contributions to enhance stereo matching performance by identifying and eliminating specular surface reflections. The identification of specular reflection can be done both passively and actively. We use dichromatic-based methods to identify and eliminate specular reflections passively. We utilize polarization imaging methods to do the same job actively. In this work we also study structured light based methods that can give better reconstruction results compared to stereo imaging methods. We propose three laser scanners equipped with a pair of line lasers and a method to calibrate these systems. Another convenient way to obtain good surface reconstruction results using structured light is to use projectors that can be used as a light source that project complicated patterns. We show our results from a digital camera-projector-based scanning system as well. This system can robustly generate a very dense reconstruction of surfaces. We also use the projector based scanning system to determine the surface reflection properties. Using high dynamic range imaging (HDRI) techniques makes it possible for us to estimate scene radiance values. Since we can determine the incoming and outgoing light directions, we are able to measure bidirectional reflectance distribution function (BRDF) values from reconstructed surface points for corresponding directions. If the sample surface have not only diffuse reflection components but also a sufficient amount of specular highlights, it is possible to approximate BRDF corresponding to a surface by fitting an analytical BRDF model to the measured data. In our work we preferred to use Phong BRDF model. Finally, we present results with rendered synthetic images where the parameter values of the Phong model were estimated using scans of real objects.