Blar i NTNU Open på forfatter "Theoharis, Theoharis"

Viser treff 21-40 av 67

    • Deep Learning on 3D Feature Descriptors 

      Puente, Philip Øygarden (Master thesis, 2020)
      I de siste årene har 3D-figur data blitt lettere å få tak i, delvis grunnet kommersielle sensorer og utbredt bruk av 3D-printere. Det er derfor ønskelig å ta i bruk dyp-læringsmetoder, som har vist seg å være anvendbare ...

    • Deep Visual Domain Adaptation: - From Synthetic Data to the Real World 

      Reiersen, Magnus (Master thesis, 2018)
      In the field of computer vision, the increasing use of convolutional neural networks (CNN) fuels the need for more and more labeled training data. Synthetic data generated from computer graphics represent an alternative ...

    • Denoising Algorithms for Ray Tracing in Real-Time Applications 

      Flatval, Håkon (Master thesis, 2020)
      Ray tracing er en bildegjengivelsesteknikk som inntil nylig var avskrevet som for beregningstung for sanntidsapplikasjoner. Teknologiske fremskritt innen grafikkmaskinvare har gjort teknikken mulig å bruke i sanntid, men ...

    • Effective Descriptors for Human Action Retrieval from 3D Mesh Sequences 

      Veinidis, Christos; Danelakis, Antonios; Pratikakis, Ioannis; Theoharis, Theoharis (Journal article; Peer reviewed, 2019)
      Two novel methods for fully unsupervised human action retrieval using 3D mesh sequences are presented. The first achieves high accuracy but is suitable for sequences consisting of clean meshes, such as artificial sequences ...

    • Efficient Sample Reusage in Path Space for Real-Time Light Transport 

      Reza, Marvin (Master thesis, 2021)
      I denne oppgaven undersøker vi ulike eksisterende løsninger for sanntidssimulering av global belysning. Nærmere bestemt, ser vi på problemstillingen rundt det å beregne mengden av direkte belysning i virtuelle scener med ...

    • Ensemble of PANORAMA-based convolutional neural networks for 3D model classification and retrieval 

      Sfikas, Konstantinos; Pratikakis, Ioannis; Theoharis, Theoharis (Journal article; Peer reviewed, 2018)
      A novel method for the classification and retrieval of 3D models is proposed; it exploits the 2D panoramic view representation of 3D models as input to an ensemble of convolutional neural networks which automatically compute ...

    • Fast and accurate GPU accelerated, high resolution 3D registration for robotic 3D reconstruction of compliant Food objects 

      Isachsen, Ulrich Johan; Theoharis, Theoharis; Misimi, Ekrem (Peer reviewed; Journal article, 2020)
      If we are to develop robust robot-based automation in primary production and processing in the agriculture and ocean space sectors, we have to develop solid vision-based perception for the robots. Accurate vision-based ...

    • Forfatting og Prosedyrisk Modellering av Terreng og Landdekke med cGANs 

      Lystrup, Asbjørn Fintland (Master thesis, 2019)
      Modellering av virtuelle landskap kan være en utfordrende og tidkrevende prosess som ofte mangler god brukerkontroll, både for menneskeforfattere og for prosedyriske modelleringsmetoder. Jeg presenterer en metode for å ...

    • Fotorealistiske bilder for objekt-gjennfinning 

      Nygård, Arve (Master thesis, 2016)
      Med basis i en database av 3d-modeller fra Princeton genererer vi et datasett med 2d-bilder som kan brukes til trening og testing av algoritmer for objekt-gjenfinning. Det resulterende datasettet samt rammeverket som ble ...

    • Fractal and multifractal analysis of PET/CT images of metastatic melanoma before and after treatment with ipilimumab 

      Breki, Christina-Marina; Dimitrakopoulou-Strauss, Antonia; Hassel, Jessica; Theoharis, Theoharis; Sachpekidis, Christos; Pan, Leyun; Provata, Astero (Peer reviewed; Journal article, 2016)
      Background PET/CT with F-18-fluorodeoxyglucose (FDG) images of patients suffering from metastatic melanoma have been analysed using fractal and multifractal analysis to assess the impact of monoclonal antibody ipilimumab ...

    • Human Pose Estimation Assisted Fitness Technique Evaluation System 

      Eivindsen, Joachim Eide; Kristensen, Brede Yabo (Master thesis, 2020)
      Vektløfting er en populær og effektiv form for styrketrening, men denne metoden kommer også med høy risiko for skade blant nye løftere. Hver person har ulike utfordringer når de skal forbedre løfte-teknikken sin og det er ...

    • Human Pose Estimation Assisted Fitness Technique Evaluation System 

      Eivindsen, Joachim Eide; Kristensen, Brede Yabo (Master thesis, 2020)
      Vektløfting er en populær og effektiv form for styrketrening, men denne metoden kommer også med høy risiko for skade blant nye løftere. Hver person har ulike utfordringer når de skal forbedre løfte-teknikken sin og det er ...

    • Image-based Somatotype as a Biometric Trait for Non-Collaborative Person Recognition at a Distance and On-The-Move 

      Danelakis, Antonios; Theoharis, Theoharis (Peer reviewed; Journal article, 2020)
      It has recently been shown in Re-Identification (Re-ID) work that full-body images of people reveal their somatotype, even after change in apparel. A significant advantage of this biometric trait is that it can easily be ...

    • Image-space Ambient Obscurance in WebGL 

      Gravås, Lorents Odin (Master thesis, 2013)
      Image-space approaches to ambient obscurance have become the de-facto standard for realistic ambient lighting in real-time applications. This thesis investigates the potential applicability of such approaches for a WebGL-based ...

    • Incremental loading of terrain textures 

      Fikkan, Eirik (Master thesis, 2013)
      Graphic Processing Units (GPUs) are getting more powerful and are currently capable of displaying millions of polygons at a steady frame rate. The challenge is however, to display unique and interesting data while navigating ...

    • An indexing scheme and descriptor for 3D object retrieval based on local shape querying 

      van Blokland, Bart Iver; Theoharis, Theoharis (Peer reviewed; Journal article, 2020)
      A binary descriptor indexing scheme based on Hamming distance called the Hamming tree for local shape queries is presented. A new binary clutter resistant descriptor named Quick Intersection Count Change Image (QUICCI) is ...

    • Looking beyond appearances: Synthetic training data for deep CNNs in re-identification 

      Barbosa, I.; Cristani, Marco; Caputo, Barbara; Rognhaugen, Aleksander; Theoharis, Theoharis (Journal article; Peer reviewed, 2018)
      Re-identification is generally carried out by encoding the appearance of a subject in terms of outfit, suggesting scenarios where people do not change their attire. In this paper we overcome this restriction, by proposing ...

    • MARF: The Medial Atom Ray Field object representation 

      Sundt, Peder Bergebakken; Theoharis, Theoharis (Peer reviewed; Journal article, 2023)
      We propose Medial Atom Ray Fields (MARFs), a novel neural object representation that enables accurate differentiable surface rendering with a single network evaluation per camera ray. Existing neural ray fields struggle ...

    • Merging Meshes from Different 3D Scanners 

      Kongevold, Dimitry (Master thesis, 2013)
      Creating digital representation of physical objects is becoming more and more popular as scanning technology becomes more available on the consumer marked. High precision close range scanners, based on structured light, ...

    • Mesh-based 3D face recognition using Geometric Deep learning 

      Wardeberg, Håkon (Master thesis, 2021)
      Face recognition has been a very active and challenging task in the Computer Vision field. Performing face recognition based on facial images can be tricky since images are illumination, scale, and pose variant. On the ...