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 23 for “"Generative priors"”.
-
LEVERAGING IMAGE GENERATIVE PRIORS FOR 3D EDITING
… extraction within captured NeRF models, using generative priors to fill in any gaps or occlusions within the object's structure. This was further expanded into complete scene inpainting methods, which drastically reduced the number of required user interactions to just a single click. Our …
-
Learned Generative Priors for Imaging Inverse Problems
… structure. Such properties are known as priors and the dominant paradigm over the last two decades or so has been to exploit the sparsity of natural images in a hand-crafted basis. In recent years, however, the field of machine learning, and deep learning in particular, has demonstrated …
-
Bayesian inverse problems with neural generative priors
… We present a data-driven framework where a generative neural network in the form of a Wasserstein Generative Adversarial Network (WGAN) learns a realistic prior, decoupled from the inverse problem in consideration. The method is tested on a severely undersampled two dimensional MRI analogue. …
-
Solving inverse problems with generative priors trained on corrupted data
… I investigate the use of score-based generative models and diffusion models trained on corrupted data to address the challenges of reconstructing signals from noisy and incomplete measurements. Unlike traditional machine learning techniques, which often rely on clean data, the proposed …
-
Perceptual quality assessment and enhancement of visual media using generative priors
… novel datasets, objective quality models, and generative techniques—for assessing and enhancing the perceptual quality of visual media in the HDR-UGC era. The first part of this work establishes the data and algorithmic foundations for HDR-UGC video quality assessment. We begin by introducing …
-
Advancing Visual Assets: Taming Deep Priors for Editing and Enhancement
… in deep learning have introduced powerful generative models (e.g., GANs, diffusion models) and human-aligned visual representations (e.g., VGG, DINO-v2) that offer promising capabilities for improving visual assets. Yet, directly applying these models to real-world editing and enhancement …
-
TOWARDS GENERALIZABLE AND EFFICIENT 3D SCENE RECONSTRUCTION
… view generation. Building on this strong generative capability, we leverage the prior knowledge of T2I models to facilitate 3D reconstruction under two challenging settings: sparse-view observations and "in-the-wild" scenarios. Specifically, GS-GS utilizes these generative priors to …
-
Towards generative modeling and control of 3D interactions
… focusing on the physical complementarity and generative priors required to understand how object parts fit together. This spatial awareness facilitates the transition to modeling the joint manifold of agents and objects, utilizing unified generative representations to synthesize dexterous, …
-
Building spatially intelligent machines : scalable 3D reconstruction, generation, and understanding
… 3D Learning with Geometric Principles and Generative Priors: We address data sparsity in photorealistic 3D asset creation by integrating geometric constraints with powerful generative priors learned from web-scale data. This combined approach not only bridges the gap caused by limited …
-
Set-restricted isometry for sub-Gaussian matrices and inversion of deep generative models
… space property for 0-1 matrices. Recently, deep generative neural networks are becoming increasingly popular for modeling certain classes of signals or images. In the second topic of this thesis, we study a novel algorithm called Partially Linearized Update for Generative Inversion (PLUGIn) for …
-
A sparsity-free compressed sensing theory with applications in generative model recovery
… set. We apply our results to the special case of generative priors, i.e. when x is close to the range of a Generative Neural Network (GNN) with ReLU activation functions. Also, if the GNN has random weights in the last layer, our theory allows a partial Fourier measurement matrix, thus taking the …
-
Scalable Vision-Language Alignment with Generative Models
… and action. This dissertation advances a generative perspective that treats large autoregressive language models and text-conditioned diffusion models as an alignment engine—using scalable text and visual generation to bring perception and language into register. Framed this way, I study …
-
Visual Perception and Reconstruction at Scale
… data acquisition. (3) Leveraging pre-trained priors from large-scale weak annotations, e.g., discriminative priors from self-supervised learning and generative priors from text-to-image generation, to improve the performance and robustness of models trained on limited task-specific data. Our …
-
A deep learning approach to wireless system design for channel sensing, contention & estimation
… part of this dissertation, we develop a deep generative learning framework to perform channel estimation from an insufficient number of pilots in high dimensional wireless systems. Channel estimation using generative priors assumes that the reconstructed channel lies in the range of the …
-
Learning from Hidden Signals in Visual Data
… them requires incorporating additional priors. The work in this thesis draws on two sources of priors: known physical and geometric facts, and generative priors learned from large-scale data. In the first part, the structure comes from physics and geometry. We use this structure to …
-
Generative and Spatial Modelling for Autonomous Driving under Adverse Weather
… this framework, this thesis delivers a suite of generative and spatial modelling methodologies for both data enhancement and data generation, progressing from 2D image-level processing to 3D scene-level representations. For data enhancement, I first introduce a unified image enhancement framework …
-
Deep learning approaches for imaging inverse problems with structured noise
… ratio (SBR) fluorescence imaging, and a novel generative modeling framework based on Whitened Score (WS) diffusion models for general imaging inverse problems with correlated Gaussian noise. The first part of this work introduces SBR-Net, a deep neural network trained on synthetic data …
-
Prior-Informed Visual Enhancement and 3D Reconstruction under Challenging Illumination
… or synthetic datasets, and neglect auxiliary priors available from external data or physics. As a result, they overfit to seen degradations and break under heavy noise, motion, or extreme dynamic range. Meanwhile, modern 3D scene representation techniques, such as Neural Radiance Fields (NeRF) …
-
Conditional Generation of Protein Sequence and Structure
… techniques from image processing to make them generative in a method termed “constrained Hallucination”. I apply the technique to optimize sequences such that their predicted structures contain desired functional sites on a slew of design problems ranging from epitope scaffolding, metal …
Page 1 of 2