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 65 for “"Stable Diffusion"”.
-
QRM: Quality-Aware Modulation for Stable Diffusion
Modern diffusion models used for image generation, such as Stable Diffusion 3.5 (SD3.5), rely on timestep or prompt embeddings to modulate the strength of the denoising process in each timestep. While this per-timestep modulation communicates the current noise level, it does not provide any …
-
Signature-Based Attribution of Images Generated by Stable Diffusion
… software to Generative Adversarial Nets, and now Diffusion Models. The Internet Watch Foundation and the National Center for Missing and Endangered Children have reported a consistent increase in reported AI generated child sexual assault materials (CSAM) and sextortion scams since the …
-
Mining Typical Phenotypes from Cell Images Using Stable Diffusion
This thesis studies whether diffusion-model typicality can improve cell selection for image-based morphological profiling. The project focuses on Broad Bioimage Bench mark Collection 021 (BBBC021), a Cell Painting-style benchmark with mechanism of-action (MoA) labels and strong batch effects. The …
-
Distilled Few-Shot Object Detection for Truck Equipment: Distilling a few-shot state-of-the-art model for truck equipment detection utilizing stable diffusion for data augmentation
… to a lightweight YOLOv11 student model. We use Stable Diffusion 3.5 for synthetic image generation with prompt engineering to augment the limited training data. The study evaluates the impact of this approach on accuracy and inference speed. The results show that knowledge distillation with …
-
Interpretable Concept Evolution Flow in Diffusion Models
Diffusion- and flow-based generative models can be viewed as trajectory processes that transport samples from a base distribution toward a data distribution through a sequence of local updates. Although these updates determine the final sample, their step-wise roles remain difficult to interpret …
-
Difuzijos modelių pritaikymas objektų vaizdų transformavimui į pikselinį stilių /
… the desired output image. One of them is the Stable Diffusion model, but image generation or style transformation using it is generic, limited to the classes the model was trained with. Stable Diffusion can be fine-tuned to generate images of specific objects or types and this work aims to …
-
Quantitative and Qualitative Analysis of Text-to-Image models
… like Generative Adversarial Networks (GANs), Diffusion Models, and Transformers. These models have shown they can create high-quality images from a variety of text prompts. However, a comprehensive analysis that examines both their performance and possible biases is often missing from existing …
-
Protocolo computacional para geração e avaliação de imagens de vestuário com modelos de difusão : aplicação de métricas objetivas na interpretação visual de croquis de moda
… de bordas de Canny, geração de imagens com Stable Diffusion v1.5 associado ao ControlNet, em inferência realizada no Stable Diffusion WebUI AUTOMATIC1111, e avaliação quantitativa e qualitativa dos resultados. O conjunto empírico foi composto por 24 croquis autorais de vestidos, organizados …
-
Exploring Representational Harms of Generative AI Systems upon Historically Marginalized Populations often Overlooked in Fairness Research
… Then, I examine how the text-to-image GAI tool Stable Diffusion portrays aspects of Indian culture and its various subcultures, seen through the eyes of 5 focus groups of Indian AI users, accompanied by further qualitative examination of images. This reveals the presence of several problematic …
-
Tvorba reklamního videa pomocí neuronových modelů
… částí. První část využívá generativní modely Stable Diffusion a Stable Video Diffusion pro tvorbu vizuálního obsahu a GPT-3.5 Turbo pro vytváření scénářů k reklamním videím. Druhá část je webová aplikace, která slouží uživatelům k zadávání potřebných informací pro reklamy a k zobrazení …
-
Umělá neuronová síť pro generování věrohodného paleo-artu
… forem až po moderní pokročilé modely, jako je Stable Diffusion, který je hlavním nástrojem této práce. Práce se soustředí na doučování již existujícího modelu tak, aby byl schopen generovat věrohodné obrázky na základě názvů prehistorických tvorů. V praktické části je popsán výběr vhodného …
-
Latent Space Explorations for Generative AI
… as Generative Adversarial Networks (GANs) and diffusion models, which produce high-quality, realistic outputs across various domains. These advancements rely on the ability of generative models to learn and encode complex patterns and semantic relationships within high-dimensional latent …
-
Spatial-Temporal change detection in satellite imaging-monitoring /
… innovative pipeline that integrates fine-tuned Stable Diffusion and our trained CycleGAN models is introduced to unify heterogeneous datasets merging captions, RGB images, and multispectral Sentinel-2 datathereby generating synthetic multispectral imagery.
-
Video Generation Techniques for Novel View Synthesis with Flow-Matching Transformers
… unseen camera viewpoints. Recent work, such as Stable Virtual Camera, shows that large-scale image diffusion models like Stable Diffusion can be adapted for pose-conditioned view synthesis by incorporating video-generation techniques with camera conditioning. In this thesis, we introduce MVFlow, …
-
AdaptPrompt with Diffusion Set: A Unified Framework for Generalizable Deepfake Detection
… content. However, recent developments in diffusion-based generative models and publicly available tools such as Stable Diffusion and DALL-E pose immediate challenges to existing detection techniques. Diffusion models generate photorealistic and high-resolution content with less observable …
-
Evaluating Multimodal AI Systems: A Comparative Analysis of Large Languagel Model-Based Models for Text, Image, and Video Generation
… using four models—DALL·E, Imagen, Grok, and Stable Diffusion. Text prompts were used to generate images, which were then converted back to text using image captioning models. BERTScore, METEOR, ROUGE, and BLEU were employed to assess the similarity between the original prompts and …
-
The Therapeutic Potential of AI-Generated Art in Short-Term Stress Management
… benefits of generative AI tools, such as stable diffusion,in helping reduce short-term stress, as well as compare them with traditional art methods. Thirty participants were randomly assigned to one of four condition groups based on the stress-inducing activity (without the stress-inducing …
-
Captchas in the LMM Era: A Reproducible Framework for Evaluating and Designing Defenses Against LMM-Powered Solvers
… solver performance. The framework combines diffusion-based image synthesis using Stable Diffusion and ControlNet with a structured prompt strategy that incorporates object labels, scene-based distractors, and verbose decoy answers. A benchmark dataset of 500 CAPTCHA challenges is constructed …
-
MarsAnywhere: Dataset and Cross-view Diffusion Model for Satellite-to-Ground View Synthesis with Mars Data
… using modern image synthesis techniques such as diffusion models. However, despite the public availability of satellite and ground level imagery on Mars, these techniques have yet to be applied to the domain due to difficulties in collating and processing the data into a usable form. We address …
-
Re_Imaged: Reimaging architecture through artificially intelligent generated images
… the process of AI text-to-image generating tool Stable Diffusion, the thesis argues that human brain involvement is necessary due to the lack of Stable Diffusion's ability to understand architectural drawings and elements, the ability to representing architectural depth through spaces and …
Page 1 of 4