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 61 for “"Segment-Anything-Model"”.

  1. A Mixture-of-Experts Approach to Fine-Tuning the Segment Anything Model

    … emergence of large, general-purpose foundation models has sparked significant interest in the broader machine learning community. Among the many models being released, the Segment Anything Model (SAM) has demonstrated exceptional capabilities for object segmentation in various settings. Given …

    rit Repository record for A Mixture-of-Experts Approach to Fine-Tuning the Segment Anything Model (opens in a new tab)

  2. IDCC-SAM: Automated cell counting in immunocytochemistry using the segment anything model

    … Dataset Cell Counting using the Segment Anything Model (IDCC-SAM), a pipeline that leverages a zero-shot approach to segment and count cells in immunocytochemistry cellular images using SAM under unprompted settings with no required manual labels. IDCC-SAM utilizes the Meta AI’s …

    iastate Repository record for IDCC-SAM: Automated cell counting in immunocytochemistry using the segment anything model (opens in a new tab)

  3. SAMPLS: A prompt engineering approach using Segment-Anything-Model for PLant Science research

    … called PlantSeg utilized U-Net for cell wall segmentation. U-Net is a neural network model that requires training with a large amount of manually labeled confocal images and lacks generalizability. In this research, we test a foundation model called the Segment Anything Model (SAM) to evaluate …

    vt Repository record for SAMPLS: A prompt engineering approach using Segment-Anything-Model for PLant Science research (opens in a new tab)

  4. ANALYZING STYLE TRANSFER ALGORITHMS FOR SEGMENTED IMAGES

    The recently developed Segment Anything Model has made grabbing semantically meaningful regions of an image easier than before. This will allow for new applications that build on this approach that weren't previously possible. This thesis investigates integrating the Segment Anything Model with …

    ecu Repository record for ANALYZING STYLE TRANSFER ALGORITHMS FOR SEGMENTED IMAGES (opens in a new tab)

  5. New methods to improve semantic segmentation on habitat classification

    … a modular pipeline that combines deep neural segmentation backbones with the Segment Anything Model (SAM) and domain knowledge rules (P-KESS). we (i) train CNN backbones (FCN/UNet/PSPNet/DeepLabV3) for pixelwise segmentation, (ii) use the Segment Anything Model (SAM) to obtain fine object …

    missouri Repository record for New methods to improve semantic segmentation on habitat classification (opens in a new tab)

  6. Comparación de algoritmos de segmentación heurística de imágenes de angiografía coronaria utilizando técnicas no supervisadas

    This paper addresses the problem of unsupervised segmentation of coronary vessels in angiography images, a challenge posed by noise, artifacts, and the complexity of vascular structures. Two segmentation approaches, W-Net and the Segment Anything Model (SAM), are compared, and a hybrid model is …

    rosario Repository record for Comparación de algoritmos de segmentación heurística de imágenes de angiografía coronaria utilizando técnicas no supervisadas (opens in a new tab)

  7. Detekce objektů na základě jejich přibližného tvaru

    Tato bakalářská práce se zabývá problematikou segmentace obrazu a následné detekce objektů na základě přibližného tvaru, s primárním zaměřením na kraslice. Hlavním cílem bylo prozkoumat, implementovat a experimentálně ověřit vybrané metody zpracování obrazu, od tradičních algoritmů po moderní …

    brno-tech Repository record for Detekce objektů na základě jejich přibližného tvaru (opens in a new tab)

  8. AI-Driven Pig Monitoring System: Behavior and Weight Analysis

    … we develop an automated pipeline using the Segment Anything Model (SAM) with deep learning, where our Xception-Net architecture achieves a mean absolute percentage error of 7.42%. For weight forecasting, we propose multi-input deep learning architectures combining spatial and temporal …

    vt Repository record for AI-Driven Pig Monitoring System: Behavior and Weight Analysis (opens in a new tab)

  9. Robust Detection and Coarse-to-Fine Localization of Fasteners for Automated EV Battery Disassembly

    … cascade framework integrating YOLO, the Segment Anything Model (SAM), and a novel Residual Refinement Network (RRN). The RRN explicitly learns to correct systematic geometric biases induced by specular reflections on metallic surfaces. Experiments demonstrate the RRN reduces localization …

    guelph Repository record for Robust Detection and Coarse-to-Fine Localization of Fasteners for Automated EV Battery Disassembly (opens in a new tab) Embargoed until 2027-04-28

  10. Uncertainty-aware fusion of foundation and task-specific models for cardiac MRI segmentation

    Vision foundation models, such as the Segment Anything Model (SAM), demonstrate strong zero-shot generalization but lack precision with anatomically challenging structures. In contrast, convolutional neural network (CNN)-based models achieve high accuracy on domain-specific data but struggle to …

    uoit Repository record for Uncertainty-aware fusion of foundation and task-specific models for cardiac MRI segmentation (opens in a new tab)

  11. Leveraging Foundation Models for Single-Image Food Portion and Calorie Estimation

    … approach that harnesses the power of foundation models--large-scale, pre-trained neural networks--to estimate food portion and caloric content from a single image. By leveraging advanced models like the Segment Anything Model (SAM), Contrastive Language-Image Pre-training Model (CLIP), Recognize …

    tamu Repository record for Leveraging Foundation Models for Single-Image Food Portion and Calorie Estimation (opens in a new tab) Embargoed until 2027-05-01

  12. Deep learning-based methods for segmentation and labelling of clay

    … of kaolinite particles is analysed. Conventional segmentation algorithms are used along with a deep learning-aided method, which brings novelty to the clay research area. The clay image is segmented to acquire morphological properties of the material, intended to be used to inform continuum models …

    chalmers Repository record for Deep learning-based methods for segmentation and labelling of clay (opens in a new tab)

  13. Durpių laukų topografinių vaizdų segmentavimas naudojant giliojo mokinimosi modelius /

    This thesis explores the segmentation of peat field topographic images using deep learning methods. Image segmentation is a core task in computer vision, enabling the automatic identi‑ fication and delineation of meaningful objects in images. The study evaluates both a pretrained zero‑shot model – …

    vilnius Repository record for Durpių laukų topografinių vaizdų segmentavimas naudojant giliojo mokinimosi modelius / (opens in a new tab)

  14. Semantic segmentation for automated annotation of satellite images /

    … investigates methodologies for the semantic segmentation of satellite images to achieve automated annotation, focusing on Lithuania’s different landscapes - Kėdainiai, Varėna, Klaipėda and Kaunas regions. Research evaluates the efficacy of segmentation models, specifically the Segment …

    vilnius Repository record for Semantic segmentation for automated annotation of satellite images / (opens in a new tab)

  15. New paradigms for radiological segmentation with AI foundation models: automatic prompting and interactive VR agent design

    … and interactive methods for medical image segmentation, addressing two major related hurdles: the scarcity of high-quality annotations for fully-supervised lesion segmentation and the labor/expertise-intensive nature of manual segmentation workflows. Our first contribution, YOLO-URSAM, is a …

    concordia Download PDF for New paradigms for radiological segmentation with AI foundation models: automatic prompting and interactive VR agent design (opens in a new tab)

  16. Mitigating effects of data heterogeneity in decentralized learning and addressing challenges of adapting foundation models

    … focus on the adaptation and optimization of the Segment Anything Model (SAM), a foundation model for image segmentation. We begin by tackling the challenges associated with data heterogeneity in Federated Learning (FL) and Fully Decentralized Learning (FDL). We introduce an innovative …

    iastate Repository record for Mitigating effects of data heterogeneity in decentralized learning and addressing challenges of adapting foundation models (opens in a new tab)

  17. COMPUTER VISION AND IOT BASED PLANT PHENOTYPING AND GROWTH MONITORING WITH 3D POINT CLOUDS

    … greenhouse conditions. We employed the Segment Anything Model (SAM) and FastSAM for precise segmentation of individual lettuce plants in high-density arrangements, overcoming challenges posed by overlapping foliage and complex backgrounds, and significantly outperforming traditional …

    cornell Repository record for COMPUTER VISION AND IOT BASED PLANT PHENOTYPING AND GROWTH MONITORING WITH 3D POINT CLOUDS (opens in a new tab)

  18. Design and Integration of Machine Learning-Based Vision System for Automated Power Line Inspection Using a Mobile Damping Robot

    … using traditional supervised machine learning models, including Random Forest, Multi-Layer Perceptron, and Gradient Boosting. Next, the second method provides a more detailed assessment by classifying conductors into four categories: Healthy, Minor Corrosion, Pollution-Induced Corrosion, and …

    vt Repository record for Design and Integration of Machine Learning-Based Vision System for Automated Power Line Inspection Using a Mobile Damping Robot (opens in a new tab)

  19. Neuroninio briaunų aptikimo taikymas lazerinei vektorinei projekcijai: vaizdo apdorojimo grandinės realizacija ir įvertinimas /

    … HED(Holistically-Nested Edge Detection). Both models are compared against Canny – a classic edge detection algorithm. Each pipeline at its beginning also uses SAM2(Segment Anything Model 2) for the purpose of separating trackable objects from the video background. The system is implemented as a …

    vilnius Repository record for Neuroninio briaunų aptikimo taikymas lazerinei vektorinei projekcijai: vaizdo apdorojimo grandinės realizacija ir įvertinimas / (opens in a new tab)

Page 1 of 4