Sökning: "pre-trained networks"

Visar resultat 1 - 5 av 115 uppsatser innehållade orden pre-trained networks.

1. 1. Exploring the Depth-Performance Trade-Off : Applying Torch Pruning to YOLOv8 Models for Semantic Segmentation Tasks

   Master-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Xinchen Wang; [2024]
   Nyckelord :Deep Learning; Semantic segmentation; Network optimization; Network pruning; Torch Pruning; YOLOv8; Network Depth; Djup lärning; Semantisk segmentering; Nätverksoptimering; Nätverksbeskärning; Fackelbeskärning; YOLOv8; Nätverksdjup;

   Sammanfattning : In order to comprehend the environments from different aspects, a large variety of computer vision methods are developed to detect objects, classify objects or even segment them semantically. Semantic segmentation is growing in significance due to its broad applications in fields such as robotics, environmental understanding for virtual or augmented reality, and autonomous driving. LÄS MER

3. 2. Automatic Semantic Role Labelling (SRL) in Swedish

   Master-uppsats, Göteborgs universitet/Institutionen för data- och informationsteknik

   Författare :Lucy Yang Buhr; [2023-10-05]
   Nyckelord :Semantic role labelling; SRL; transfer learning; multi-task learning; multilingual transfer; T5; mT5;

   Sammanfattning : In this paper, using deep learning networks, the first end-to-end semantic role labelling model (SRL) has been developed for Swedish texts. This Swedish SRL model can, with a given Swedish sentence, perform trigger identification, frame classification and argument extraction tasks automatically in a series. LÄS MER

4. 3. Image-classification for Brain Tumor using Pre-trained Convolutional Neural Network : Bildklassificering för hjärntumör medhjälp av förtränat konvolutionell tneuralt nätverk

   M1-uppsats, KTH/Hälsoinformatik och logistik

   Författare :Ahmad Osman; Bushra Alsabbagh; [2023]
   Nyckelord :Brain tumor; Deep learning; Convolutional Neural Network CNN ; diagnosis; Image classification; pre-trained models; dataset; economic impact.; Cancer; Hjärntumör; Artificiell intelligens AI ; djupinlärning; konvolutionellt neuralt nätverk CNN ; Diagnostik; Bildklassificering; förtränade modeller; dataset.;

   Sammanfattning : Brain tumor is a disease characterized by uncontrolled growth of abnormal cells inthe brain. The brain is responsible for regulating the functions of all other organs,hence, any atypical growth of cells in the brain can have severe implications for itsfunctions. The number of global mortality in 2020 led by cancerous brains was estimatedat 251,329. LÄS MER

5. 4. Image-classification for Brain Tumor using Pre-trained Convolutional Neural Network

   Kandidat-uppsats, KTH/Skolan för elektroteknik och datavetenskap (EECS)

   Författare :Bushra Alsabbagh; [2023]
   Nyckelord :Brain tumor; Deep learning; Convolutional Neural Network CNN ; diagnosis; Image classification; pre-trained models; dataset; economic impact.; Cancer; Hjärntumör; Artificiell intelligens AI ; djupinlärning; konvolutionellt neuralt nätverk CNN ; Diagnostik; Bildklassificering; förtränade modeller; dataset.;

   Sammanfattning : Brain tumor is a disease characterized by uncontrolled growth of abnormal cells in the brain. The brain is responsible for regulating the functions of all other organs, hence, any atypical growth of cells in the brain can have severe implications for its functions. LÄS MER

7. 5. Deep Learning-Based Depth Estimation Models with Monocular SLAM : Impacts of Pure Rotational Movements on Scale Drift and Robustness

   Master-uppsats, Linköpings universitet/Datorseende

   Författare :Daniel Bladh; [2023]
   Nyckelord :Deep Learning; Computer Vision; Monocular; SLAM; Depth Estimation;

   Sammanfattning : This thesis explores the integration of deep learning-based depth estimation models with the ORB-SLAM3 framework to address challenges in monocular Simultaneous Localization and Mapping (SLAM), particularly focusing on pure rotational movements. The study investigates the viability of using pre-trained generic depth estimation networks, and hybrid combinations of these networks, to replace traditional depth sensors and improve scale accuracy in SLAM systems. LÄS MER