Sökning: "3D shape recognition"

Visar resultat 1 - 5 av 7 uppsatser innehållade orden 3D shape recognition.

  1. 1. Meta-Pseudo Labelled Multi-View 3D Shape Recognition

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

    Författare :Fehmi Ayberk Uçkun; [2023]
    Nyckelord :3D shape recognition; 3D object classification; 3D shape retrieval; 3D object retrieval; Automatic labelling; Semi-supervised learning; Pseudo labelling; Meta Pseudo Labelling; Multi-View Convolutional Neural Networks; Shape descriptors; Multi-view representations; Deeplearning; 3D-formigenkänning; 3D-objektklassificering; 3D-formhämtning; Hämtning av 3D-objekt; Automatisk märkning; Halv-vägledd lärning; Pseudomärkning; Meta Pseudo-märkning; Multi-View Faltningsnät; Formbeskrivningar; Multi-view representation; Djupinlärning;

    Sammanfattning : The field of computer vision has long pursued the challenge of understanding the three-dimensional world. This endeavour is further fuelled by the increasing demand for technologies that rely on accurate perception of the 3D environment such as autonomous driving and augmented reality. LÄS MER

  3. 2. 3D Shape Detection for Augmented Reality

    Uppsats för yrkesexamina på avancerad nivå, KTH/Skolan för elektroteknik och datavetenskap (EECS)

    Författare :Hector Anadon Leon; [2018]
    Nyckelord :3D Machine Learning; object detection; computer vision; shape; point cloud;

    Sammanfattning : In previous work, 2D object recognition has shown exceptional results. However, it is not possible to sense the environment spatial information, where the objects are and what they are. LÄS MER

  4. 3. Data Augmentation by Using Object Shape Reconstruction

    Master-uppsats, Lunds universitet/Matematik LTH

    Författare :Mats Granholm; Filip Johannesson; [2017]
    Nyckelord :Morphable models; Generative adversarial networks; Data augmentation; Mathematics and Statistics; Technology and Engineering;

    Sammanfattning : Achieving high performance face recognition often requires large manually labeled training datasets. As such datasets can be difficult to obtain, we investigate whether smaller datasets can be augmented synthetically in order to increase performance. We use 3D morphable models to create 3D reconstructions of faces from only a single image. LÄS MER

  5. 4. Selfie to Avatar - Creating a 3D Model of a Face from Front and Side View Photos

    Master-uppsats, Lunds universitet/Matematik LTH

    Författare :Mikael Högström; [2016]
    Nyckelord :image analysis; active shape models; automatic avatar creation; 3D graphics; Mathematics and Statistics;

    Sammanfattning : In this thesis we present a system for creating a textured, animated 3D model of a user’s face in real time. The model, which we call an avatar, is reconstructed from two photos of a subject, one from a front view and one from a side view. Each photo is segmented using Active Shape Models. LÄS MER

  7. 5. Matching of geometrically and topologically changing meshes

    Uppsats för yrkesexamina på avancerad nivå, Uppsala universitet/Avdelningen för beräkningsvetenskap

    Författare :Kristoffer Jonsson; [2015]
    Nyckelord :mesh; meshes; matching; matchings; topology; geometry; manifold; compression; animation; animations; fluid; spherical harmonics; tensor; bijective; game; games; DICE;

    Sammanfattning : The aim for this thesis is to develop a foundation for a compression system for animated mesh sequences, specifically under dynamic change of mesh geometry and topology. Compression of mesh sequences is of special interest in the game industry and this particular thesis is a part of an ongoing series of projects at EA DICE. LÄS MER