Sökning: "3D Human Pose Estimation"

Visar resultat 1 - 5 av 8 uppsatser innehållade orden 3D Human Pose Estimation.

  1. 1. Unsupervised 3D Human Pose Estimation

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

    Författare :Sri Datta Budaraju; [2021]
    Nyckelord :Computer Vision; Projective Geometry; Deep Learning; Unsupervised Learning; 3D Human Pose Estimation; GAN; AutoEncoder; Hybrid Generative Model; Self Supervision;

    Sammanfattning : The thesis proposes an unsupervised representation learning method to predict 3D human pose from a 2D skeleton via a VAEGAN (Variational Autoencoder Generative Adversarial Network) hybrid network. The method learns to lift poses from 2D to 3D using selfsupervision and adversarial learning techniques. LÄS MER

  2. 2. 3D Human Pose and Shape-aware Modelling

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

    Författare :Polydefkis Gkagkos; [2020]
    Nyckelord :;

    Sammanfattning : The focus of this thesis is the task of 3D pose estimation while taking into consideration the shape of a person in a single image. For rendering the human pose and the body shape we use a newly proposed statistical model, the SMPL [1]. We train a neural network to estimate the shape and the pose of a person in an image. LÄS MER

  3. 3. Real Time Gym Activity Detection using Monocular RGB Camera

    Master-uppsats, Högskolan i Halmstad/Akademin för informationsteknologi

    Författare :Mohammad Samer Alshatta; [2020]
    Nyckelord :ActivityDetection; RNN; GymActionDetection; PoseBasedActionDetection; DeepLearning; SkeletonBasedOnlineActionDetection;

    Sammanfattning : Action detection is an attractive area for researchers in computer vision, healthcare, physiotherapy, psychology, and others. Intensive work has been done in this area due to its wide range of applications such as security surveillance, video tagging, Human-Computer Interaction (HCI), robotics, medical diagnosis, sports analysis, interactive gaming, and many others. LÄS MER

  4. 4. En jämförelse av inlärningsbaserade lösningar för mänsklig positionsuppskattning i 3D

    M1-uppsats, KTH/Hälsoinformatik och logistik; KTH/Hälsoinformatik och logistik

    Författare :Alfons Lange; Erik Lindfors; [2019]
    Nyckelord :pose estimation; body pose; 3D; RGB image; Microsoft Kinect; positionsuppskattning; kroppsposition; 3D; RGB-bild; Microsoft Kinect;

    Sammanfattning : Inom områden som idrottsvetenskap och underhållning kan det finnas behov av att analysera en människas kroppsposition i 3D. Dessa behov kan innefatta att analysera en golfsving eller att möjliggöra mänsklig interaktion med spel. LÄS MER

  5. 5. Robust Localization of Research Concept Vehicle (RCV) in Large Scale Environment

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

    Författare :Anchit Raghuram; [2018]
    Nyckelord :Localization; Normal Distribution Transform; Pose Fusion; MCL; Monte-Carlo Localization; GPS; Robust Localization; SLAM;

    Sammanfattning : Autonomous vehicles in the recent era are robust vehicles that have the capability to drive themselves without human involvement using sensors and Simultaneous Localization and Mapping algorithms, which helps the vehicle gain an understanding of its environment while driving with the help of laser scanners (Velodyne), IMU and GPS to collect data and solidify the foundation for locating itself in an unknown environment. Various methods were studied and have been tested for increasing the efficiency of registration and optimization over the years but the implementation of the NDT library for mapping and localization have been found to be fast and more accurate as compared to conventional methods. LÄS MER