Global ETD Search
Search theses and dissertations gathered from participating repositories worldwide. Every result links back to the library that holds it. No account is needed.
Results
Showing 1 to 16 of 16 for “"3D Classification"”.
-
3D Classification of Power Line Scene Using Airborne Lidar Data
… scene, which are mostly based on 2-dimensional classification. Thus, the objective of this dissertation is to develop a 3-dimensional classification system which is able to automatically identify key objects in the power line corridor from large-scale LiDAR data. This dissertation introduces new …
-
Machine learning methods for 3D object classification and segmentation
… especially in the context of images. However, 3D object understanding is still not on par with its 2D domain and deep learning for 3D has not been fully explored yet. In this dissertation, I work on two approaches, both of which advances the state-of-the-art results in 3D classification and …
-
An Empirical Study On Sampling Approaches For 3D Image Classification Using Deep Learning
A 3D classification method requires more training data than a 2D image classification method to achieve good performance. These training data usually come in the form of multiple 2D images (e.g., slices in a CT scan) or point clouds (e.g., 3D CAD modeling) for volumetric object representation. The …
-
Vision Language Model Guided Zero-shot Classification
… core tasks in Computer Vision, 2D image and 3D object classification are fundamental tasks which serve as the foundation for numerous applications including scene understanding, robotics and autonomous navigation. Vision Language Model (VLMs) are deep learning architectures designed to …
-
Computer Aided Analysis of Late Gadolinium Enhanced Cardiac MRI
… values. In this thesis, we present a complete 3D framework for automatic analysis of LGE CMR images. This framework achieves 3D segmentation of the myocardium as well as 3D classification of infarcts within the segmented myocardium, with robust and effective preprocessing measures overcoming …
-
Robust 4D Millimeter-Wave Radar Perception in Adversarial Environments
… This design enhances compatibility with modern 3D perception architectures, reduces computational overhead, and enables seamless integration within existing autonomous driving stacks. The study begins with a comprehensive analysis of perception sensing modalities---camera, lidar, and radar---to …
-
Towards Domain Robustness and Adversarial Robustness of 3D Point Cloud Perception Model
Despite the recent success of deep learning-based 3D point cloud perception models, the domain robustness and adversarial robustness of such models raised concerns. For domain robustness, point cloud data can be captured with different sensors, which produce different 3D sampling patterns. A …
-
Determining the Structure of Glial Fibrillary Acidic Protein (GFAP) Filaments by Cryo-Electron Microscopy and Tomography
… the helical parameters of GFAP and obtained a 3D reconstruction of the GFAP filament at 7.4 Å resolution. Extensive 3D classification and in silico filament reconstitution revealed that GFAP filaments exhibit continuous helical heterogeneity along their length. Next, I constructed a molecular …
-
Block Pattern adaptation for Greek female adolescents with Scoliosis of the Spine: An investigation into the feasibility of incorporating body shape asymmetry into Sizing Systems to improve garment fit
… and degrees of spinal curvature, providing a 3D classification of scoliotic deformities. Both right and left body halves of 75 females aged 16-22 years of age, diagnosed with Adolescent Idiopathic Scoliosis (AIS), are measured in order to register their different body shapes and to classify …
-
Phytolith-Based Paleoecology and Hominin Landscapes at Oldupai Gorge 2–1.4 Ma
… analysis was done through fresh mounting for 3D classification. Quantitative analysis was done using IBM® SPSS®, while Principal Component Analysis (PCA) used PAleontological STatistics (PAST) software. The focus was on examining the differences and similarities between the two zones (Eastern …
-
Sensor Fusion for Precise Mapping of Transparent and Specular Reflective Objects
… requirements it has to be able to build a 3D map of the environment. This is essential to navigate through rough terrain and fulfil manipulation tasks (e.g. open doors). To build a map and gather environmental information, robots are equipped with multiple sensors. Since laser scanners …
-
Structural Determination of HER2 in Its Native Membrane Environment
… single particle analysis (SPA) from cryo-EM. STA 3D classification proved difficult due to the lower signal to noise ratio of subtomograms and the limited information caused by the ‘missing wedge’ effect. SPA proved to be the more suitable method to reconstruct the structure of the receptor. …
-
Structural studies of the assembly of Vibrio sheathed polar flagellum
… used to determine the structure. After 2D and 3D classification, high-resolution structures of sheathed and sheath-free <i>Vibrio</i> filaments were obtained. These structures reveal that the <i>Vibrio</i> filament sheath is the outer membrane, highly modified by lipopolysaccharides (LPS), and …
-
Structural basis of translational recycling and bacterial ribosome rescue
… initiation complexes were solved and intensive 3D classification allowed to distinguish different stages of initiation, during which ABCE1 may play a role. Surprisingly, ABCE1 adopted a previously unknown state for ABC-type ATPases that was termed “hybrid state”. Here, the NBSI is in a half open …
-
Very High Resolution & 3D Remote Sensing Data for Supporting Forestry in Wallonia. What Resolution? What Scale? What Purpose?
The forest, which represents one-third of the Earth surface, has always been in close interaction with human societies. Forests provide many environmental services: soil and water resources protection, reduction of impact of gas emissions, biodiversity reservoir and conservation... Concerns about …