Sökning: "självövervakat lärande"

Hittade 3 uppsatser innehållade orden självövervakat lärande.

  1. 1. Evaluating the effects of data augmentations for specific latent features : Using self-supervised learning

    M1-uppsats, KTH/Hälsoinformatik och logistik

    Författare :Markus Ingemarsson; Jacob Henningsson; [2022]
    Nyckelord :Contrastive learning; data augmentations; deep learning; invariant features; machine learning; representation similarity analysis; self-supervised learning; SimCLR; Kontrast inlärning; datamodifieringar; djupinlärning; maskininlärning; SimCLR; självövervakat lärande; oföränderliga egenskaper; representativ likhetsanalys;

    Sammanfattning : Supervised learning requires labeled data which is cumbersome to produce, making it costly and time-consuming. SimCLR is a self-supervising framework that uses data augmentations to learn without labels. This thesis investigates how well cropping and color distorting augmentations work for two datasets, MPI3D and Causal3DIdent. LÄS MER

  3. 2. Self-supervised Learning for Efficient Object Detection

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

    Författare :Benjamin István Berta; [2021]
    Nyckelord :Self-supervised Learning; Object Detection; Computer Vision; Contrastive Learning; Deep Learning; Självövervakat lärande; Objektdetektering; Datorsyn; Contrastive Learning; Deep Learning;

    Sammanfattning : Self-supervised learning has become a prominent approach in pre-training Convolutional Neural Networks for computer vision. These methods are able to achieve state-of-the-art representation learning with unlabeled datasets. In this thesis, we apply Self-supervised Learning to the object detection problem. LÄS MER

  4. 3. Self-supervised learning of camera egomotion using epipolar geometry

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

    Författare :Fereidoon Zangeneh Kamali; [2020]
    Nyckelord :;

    Sammanfattning : Visual odometry is one of the prevalent techniques for the positioning of autonomous agents equipped with cameras. Several recent works in this field have in various ways attempted to exploit the capabilities of deep neural networks to improve the performance of visual odometry solutions. LÄS MER