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 20 of 30 for “"3D Deep Learning"”.
-
Scalable 3D deep learning: methods and applications
In recent years, deep convolutional neural networks (CNNs) have achieved remarkable success in most 2D image processing tasks. Still, the direct processing of 3D data, not only their 2D projections, is required in many applications. Naive extensions of 2D deep learning techniques into 3D rely on …
-
3D Deep Learning for Object-Centric Geometric Perception
… aims at extracting the geometric attributes of 3D objects. These attributes include shape, pose, and motion of the target objects, which enable fine-grained object-level understanding for various tasks in graphics, computer vision, and robotics. With the growth of 3D geometry data and 3D deep …
-
Circuit architecture co-design for machine learning and 3D deep learning
… the same time, new scene representations such as 3D Gaussian Splatting (3DGS) enable real-time photorealistic rendering and efficient spatial reasoning for immersive applications including robotics, autonomous systems, and AR/VR. These algorithms provide unprecedented capabilities but also …
-
3D Deep Learning for Pulmonary Nodule Malignancy Classification in CT Imaging
… variability. This thesis proposes a reproducible deep learning framework for pulmonary nodule malignancy classification using volumetric CT data. Using the LIDC-IDRI dataset, a standardized preprocessing pipeline is developed, including isotropic resampling, intensity normalization, and 3D patch …
-
A 3D Deep Learning Architecture for Denoising Low-Dose CT Scans
This paper introduces 3D-DDnet, a cutting-edge 3D deep learning (DL) framework designed to improve the image quality of low-dose computed tomography (LDCT) scans. Although LDCT scans are advantageous for reducing radiation exposure, they inherently suffer from reduced image quality. Our novel 3D DL …
-
BRIDGING INTERPRETABILITY AND PERFORMANCE IN 3D DEEP LEARNING THROUGH GEOMETRIC INDUCTIVE BIASES
… possano risolvere limitazioni fondamentali nel deep learning 3D, in particolare per applicazioni critiche di sicurezza come l’ispezione delle reti elettriche. Nonostante i progressi significativi nel campo del 3D scene understanding, la maggior parte dei modelli allo stato dell’arte funziona …
-
H3DNET: A Deep Learning Framework for Hierarchical 3D Object Classification
Deep learning has received a lot of attention in the fields such as speech recognition and image classification because of the ability to learn multiple levels of features from raw data. However, 3D deep learning is relatively new but in high demand with their great research values. Current …
-
ON ADVERSARIAL MACHINE LEARNING AND ROBUST OPTIMIZATION
This thesis considers adversarial machine learning and robust optimization problems. Adversarial machine learning studies the robustness and security facet of machine learning algorithms, in particular deep learning, while robust optimization is a powerful technique to tackle uncertainty in …
-
Deep Learning for Infinite Virtual Urban Environment Generation
… in games or Virtual Reality applications in 3D. Deep Learning technologies, which have advanced in various scientific research fields at an extremely high speed, have brought new possibilities to generate the virtual urban environment. Our research is concentrated on generating virtual urban …
-
3D OBJECT DETECTION AND POSE ESTIMATION USING POINT CLOUDS
… need for advanced algorithms that are useful for 3D data analysis and interpretation. This thesis studies the problem of 3D object detection for point clouds in indoor environments, focusing on heavily cluttered and occluded objects. While 3D deep learning networks are split between methods that …
-
Integrating geometric and physics priors into 3D shape generation and deformation
3D geometry is a cornerstone of 3D computer vision, offering invaluable prior knowledge for advancing 3D deep learning. While current deep learning frameworks often focus on capturing the probabilistic distribution from large datasets, this proposal highlights how integrating geometric priors and …
-
Cybergis-enabled remote sensing data analytics for deep learning of landscape patterns and dynamics
… and better understand landscape dynamics from a 3D perspective. LiDAR data have been applied to diverse remote sensing applications where large-scale landscape mapping is among the most important topics. While researchers have used LiDAR for understanding landscape patterns and dynamics in many …
-
Three-Dimensional Medical Image Registration with Applications in Proton Therapy
… or accuracy. We enhance DIR methods using deep-learning-based approaches, with a specific emphasis on improving registration accuracy and error estimation critical for clinical applications. We introduce a 3D deep-learning-based multi-modality image registration correction network (RCN) …
-
Predictions of Periventricular Fazekas Scores on MRI: Addressing Label Noise, Interpretability and Uncertainty
… Predicting Fazekas scores directly using deep learning could provide a scalable alternative to the current WMH segmentation analysis. However, this direction is underexplored, probably due to limited model performance and the lack of inherent explainability in classification tasks. This …
-
Deep learning for brain structure and function analysis from neuroimaging data
… and psychological conditions. Applying deep learning to this domain allows for the analysis of complex neuroimaging data with unprecedented accuracy and efficiency. This thesis utilises deep learning to explore three problems concerning brain structure and function: white matter …
-
Exploration of Deep Learning Applications on an Autonomous Embedded Platform (Bluebox 2.0)
… using the Artificial neural network or Deep neural network, as an imitation of the traditionally used algorithms. Recent research in the field of deep learning and development of competent processors for autonomous or self-driving car have shown amplitude of prospect, but there are many …
-
Robust and generalizable deep geometric representation learning for 3D point clouds
Deep learning has revolutionized 3D perception, enabling accurate recognition, segmentation, and reconstruction of point-cloud data across robotics, autonomous navigation, and augmented reality. However, two major challenges remain. First, large-scale 3D models rely heavily on labeled datasets, …
-
Deep Learning for Improved Quantification of Uptake in Hybrid Imaging
… of oncologic patients. In this thesis, a 3D deep learning approach is developed with the aim of improving quantification of tumour radiotracer activity concentration and tumour shape definition in PET images. A convolutional neural network (CNN) is trained on a set of simulated \ground …
-
3D diffusion based generation model for point cloud annotation and generation
… with the rapid advancement of applications and 3D scanning sensors, the demand for 3D deep learning based technology and data has increased dramatically. Especially 3D shape with semantic labels plays a significant role in 3D vision problems, such as auto-driven, 3D object detection and 3D scene …
Page 1 of 2