Hello!

Welcome to RobertoDyke.com

About

I am currently a research engineer on the Cloud Platform For Smart City (CP4SC) project at Inria's centre in Sophia Antipolis, France. My work focuses on developing reconstruction algorithms of buildings from point clouds.

Résumé

A current copy of my résumé is available as a PDF.

Publications

2022

  • article Histogram equalization using a selective filter

    Roberto M. Dyke, Kai Hormann
    The Visual Computer, 2022
    Paper

  • article A Survey of Non-Rigid 3D Registration

    Bailin Deng, Yuxin Yao, Roberto M. Dyke, Juyong Zhang
    Eurographics STAR - State of The Art Report (presented at EG), 2022
    Paper · arXiv Preprint

2021

  • article Registration of non-rigidly deforming objects

    Roberto M. Dyke
    School of Computer Science & Informatics, Cardiff University, 2021
    Project page · Thesis

2020

  • accessibility SHREC’20: Shape correspondence with non-isometric deformations

    Roberto M. Dyke, Yu-Kun Lai, Paul L. Rosin, Stefano Zappalà, Seana Dykes, Daoliang Guo, Kun Li, Riccardo Marin, Simone Melzi, Jingyu Yang
    Computers & Graphics (presented at 3DOR), 2020
    Project page · Paper

  • accessibility SHREC'20: Non-rigid Shape Correspondence of Physically-Based Deformations

    Roberto M. Dyke, Feng Zhou, Yu-Kun Lai, Paul L. Rosin, Daoliang Guo, Kun Li, Riccardo Marin, Jingyu Yang
    3D Object Retrieval, 2020
    Project page · Paper

2019

  • article Non-rigid registration under anisotropic deformations

    Roberto M. Dyke, Yu-Kun Lai, Paul L. Rosin, Gary K. L. Tam
    Computer Aided Geometric Design (presented at GMP), 2019
    Paper

  • accessibility SHREC'19: Shape Correspondence with Isometric and Non-Isometric Deformations

    Roberto M. Dyke, C. Stride, Yu-Kun Lai, Paul L. Rosin, Mathieu Aubry, Amit Boyarski, Alexander M. Bronstein, Michael M. Bronstein, Daniel Cremers, Matthew Fisher, Thibault Groueix, Daoliang Guo, Vladimir G. Kim, Ron Kimmel, Zorah Lähner, Kun Li, Or Litany, Tal Remez, Emanuele Rodolà, Bryan C. Russell, Yusuf Sahillioğlu, Ron Slossberg, Gary K. L. Tam, Matthias Vestner, Z. Wu, Jingyu Yang
    3D Object Retrieval, 2020
    Project page · Paper

2018

  • article HDFD — A High Deformation Facial Dynamics Benchmark for Evaluation of Non-Rigid Surface Registration and Classification

    Gareth Andrews, Sam Endean, Roberto M. Dyke, Yu-Kun Lai, Gwenno Ffrancon, Gary K. L. Tam
    arXiv Preprint

Other activities

The source code of my Cloud Compare plugin for point cloud classification is available on Github. The implementation uses a random forest classifier, which requires very little training data yet achieves good results.

I had the pleasure of helping to organise the 3DOR 2021 workshop.

Datasets

During the course of my PhD I created several datasets for the evaluation of non-rigid registration algorithms:

SHREC 2019

A tiered dataset of progressively complex groups of shapes, ranging from isometric to non-isometric deformations.

External link
SHREC 2020a

A dataset of highly complex non-isometric non-rigid deformations.

Link
SHREC 2020b

A dataset of non-rigid quadruped shapes with consensus-based ground-truth correspondences.

Link