Patrik Goorts | Site / AllRecentChanges

Main / Publications

2020-04-29T07:49:19Z

2016

New visual coding exploration in MPEG: Super-MultiView and Free Navigation in Free viewpoint TV
Multi-view wide baseline depth estimation robust to sparse input sampling
Omnidirectional Free Viewpoint Video using Panoramic Light Fields
Nonuniform Depth Distribution Selection with Discrete Fourier Transform

2015

Real-Time Edge-Sensitive Local Stereo Matching with Iterative Disparity Refinement
Automatic Calibration of Soccer Scenes Using Feature Detection
Multi-Camera Epipolar Plane Image Feature Detection for Robust View Synthesis

2014

Optimal Distribution of Computational Power in Free Viewpoint Interpolation by Depth Hypothesis Density Adaptation in Plane Sweeping (Paper)
Iterative Refinement for Real-time Local Stereo Matching (Paper)
Free Viewpoint Video for Soccer using Histogram-Based Validity Maps in Plane Sweeping (Details)
A qualitative comparison of MPEG view synthesis and light field rendering (Paper)
Self-Calibration of Large Scale Camera Networks (Paper)
Tile Tracker: A Practical and Inexpensive Positioning System for Mobile AR Applications (Paper) (Poster)
Real-time Local Stereo Matching using Edge Sensitive Adaptive Windows (Paper)
Plane Sweeping in Eye-gaze Corrected, Tele-immersive 3D Video Conferencing (Paper)

2013

Bringing 3D vision to the web: Acquiring motion parallax using commodity cameras and WebGL (Details)
Optimization of Free Viewpoint Interpolation by Applying Adaptive Depth Plane Distributions in Plane Sweeping (Details)
Scalable optical tracking for navigating large virtual environments using spatially encoded markers (Details)
GPU-based View Interpolation for Smooth Camera Transitions in Soccer (Details)
Real-time Video-Based View Interpolation of Soccer Events using Depth-Selective Plane Sweeping (Details)

2012

An End-to-end system for Free Viewpoint Video for Smooth Camera Transitions (Details)
Raw Camera Image Demosaicing using Finite Impulse Response Filtering on Commodity GPU Hardware using CUDA (Details)
ARDO: Automatic Removal of Dynamic Objects (Details)
HeatMeUp: a 3DUI Serious Game to Explore Collaborative Wayfinding (Details)

2011

Look Mother, Virtual Puzzling Without Buttons! (Details)

2010

Practical examples of GPU computing optimization principles (Details)

2009

Optimal Data Distribution for Versatile Finite Impulse Response Filtering on Next-Generation Graphics Hardware using CUDA (Details)

Thesis

Bachelorthesis - Moving a mouse without a mouse (2007) (Details)
Masterthesis - Mooie resultaten met weinig moeite: Dieptemappen met stroke-based invoer (2009) (Details)

Bibliography

Bibtex
PDF

Main / Projects

2020-04-29T07:44:32Z

Volumetric reconstruction for holographic communication and collaboration in AR

This project is realized under Magic Leap

Free viewpoint video for smooth camera transitions for soccer games

We propose a project in the context of the broadcasting of sport events. We present a new camera system which allows broadcasting of sport events from different angles, currently impractical or impossible. This is accomplished by using a virtual camera, meaning that the user can control a camera that does not exist in the real world. The user can choose a position of the virtual camera and a computer can then generate the image of that virtual camera, just as if there would be a real camera at that position.

This virtual camera system allows a number of novel applications. For example, the game can be frozen and shown from an arbitrary number of angles. This is currently impossible, because a real camera can only be at one place at a time. Furthermore, many angles can be broadcasted, which are currently unavailable due to the placement of the cameras. This gives broadcasters better tools for better television.

This virtual camera system is demonstrated for soccer and is presented in a PhD thesis. All results, text, images, and videos of the PhD can be found here: http://patrikgoorts.com/phd

Project website: http://projectfine.eu. UHasselt is developing only the interpolation part.

Results

https://www.youtube.com/watch?v=6MzeXeavE1s

https://www.youtube.com/watch?v=y_jeFam1p5c

Setup and Method

See the phd text and papers for more information:

PhD (233 MiB, pdf)

Camera setup

System setup

Rendering method

Main / EducationAndWork

2020-04-28T16:00:37Z

2019-present: Senior researcher at Magic Leap (link)
2009-2019: PhD student and researcher (Computer Science) @ EDM Hasselt (link)
2007-2009: Master computer science @ University Hasselt (link)
2004-2007: Bachelor computer science @ University Hasselt (link)
1998-2004: Science & Mathematics @ KA1 Hasselt (link)

Main / Contact

2020-04-28T15:58:52Z

If you want to contact me, please reach out on linkedin

Facebook: http://patrikgoorts.com/facebook
LinkedIn: http://patrikgoorts.com/linkedin
UHasselt: http://patrikgoorts.com/work

Main / Education

2020-04-28T15:57:56Z

2009-present: PhD student (Computer Science) @ EDM Hasselt (link)
2007-2009: Master computer science @ University Hasselt (link)
2004-2007: Bachelor computer science @ University Hasselt (link)
1998-2004: Science & Mathematics @ KA1 Hasselt (link)

Publications / Webgl3d

2014-03-04T13:47:24Z

Bringing 3D vision to the web: Acquiring motion parallax using commodity cameras and WebGL

Details

Authors: Patrik Goorts, Steven Maesen, Dimitri Scarlino, Philippe Bekaert

Language: English

Year: 2013

Journal: Proceedings of the International Conference on 3D Imaging (IC3D 2013)

Link: http://patrikgoorts.com/Publications/goorts2013bringing.pdf

Link: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6732092

Images

Bibtex

@inproceedings{goorts2013bringing,
	title={Bringing 3D vision to the web: Acquiring motion parallax using commodity cameras and WebGL},
	author={Goorts, P. and Maesen, S. and Scarlino, D. and Bekaert, P.},
	booktitle={Proceedings of the International Conference on 3D Imaging (IC3D 2013)},
	year={2013},
	organization={IEEE}
}

Abstract

In this paper, we present a system to acquire 3D vision by motion parallax on web-based platforms using head tracking. We employ the camshift algorithm to perform color-based head tracking. Using the position of the head, we can render a 3D scene from the viewpoint of the viewer, thus acquiring motion parallax, a strong cue for 3D vision. We employed web technologies to allow the adoption of our method to any modern device, including mobile devices. WebGL is used for rendering and head tracking, and WebRTC is used for camera input. No software installation or plugins are required. We demonstrated the effectiveness of our method on a variety of devices, such as desktop computers, laptops, and tablets.

Publications / 3DUI-2011

2014-03-04T13:41:45Z

Look Mother, Virtual Puzzling without Buttons!

Details

Authors: Steven Maesen, Patrik Goorts, Lode Vanacken, Sofie Notelaers, Tom De Weyer

Language: English

Year: 2011

Journal: IEEE Symposium on 3D User Interfaces (3DUI), 2011

Link: http://patrikgoorts.com/Publications/maesen2011look.pdf

Videos

https://www.youtube.com/watch?v=LsZzVxidfFY

https://www.youtube.com/watch?v=SId-wFTBQbs

Bibtex

@inproceedings{Maesen3DUI11,
author = {Steven Maesen and Patrik Goorts and Lode Vanacken and Sofie Notelaers and Tom De Weyer},
title = {Look Mother, Virtual Puzzling without Buttons!},
booktitle = {2011 IEEE Symposium on 3D User Interfaces (3DUI 2011)},
pages = {139–140},
year = {2011},
month = {March},
address = {Singapore},
organization = {IEEE},
isbn = {978-1-4577-0037-8}
}

Abstract

Not many virtual 3D puzzle applications have been researched. We present a solution using tangible objects and a reach-in set-up. The user can solve a puzzle using natural and tangible interaction. To provide realistic interaction, the usage of buttons was avoided by design. In a user experiment the solution proved to be natural and intuitive.

Publications / 3DUI-2012

2014-03-04T13:40:38Z

HeatMeUp: a 3DUI Serious Game to Explore Collaborative Wayfinding

Details

Authors: Sofie Notelaers, Tom De Weyer, Patrik Goorts, Steven Maesen, Lode Vanacken, Karin Coninx, Philippe Bekaert

Language: English

Year: 2012

Journal: IEEE Symposium on 3D User Interfaces (3DUI), 2012

Link: http://patrikgoorts.com/Publications/notelaers2012heat.pdf

IEEE: http://ieeexplore.ieee.org/xpl/articleDetails.jsp?arnumber=6184219

Images

Video

https://www.youtube.com/watch?v=zz1uL6Zt4FI

Bibtex

@inproceedings{notelaers2012heat,
author = {Sofie Notelaers and Tom De Weyer and Patrik Goorts and Steven Maesen and Lode Vanacken and Karin Coninx and Philippe Bekaert},
title = {HeatMeUp: a 3DUI Serious Game to Explore Collaborative Wayfinding},
booktitle = {2012 IEEE Symposium on 3D User Interfaces (3DUI 2012)},
pages = {177–178},
year = {2012},
month = {March},
address = {Orange County, California},
organization = {IEEE},
isbn = {978-1-4673-1204-2}
}

Abstract

Wayfinding inside a virtual environment is a cognitive process during navigation. Normally the user inside the virtual environment has to rely on himself and different cues such as waypoints to improve his knowledge with regard to his surroundings. In this paper we will present our solution for the 3DUI Contest 2012: HeatMeUP, a 3DUI serious game to explore collaborative alternatives, in which a partner is responsible for providing wayfinding cues. The game is set in a multi-storey building where several fires and gas leaks occur and a firefighter has to overcome several challenges, guided by a fire chief.

Publications / Interpolation1-2014

2014-03-04T13:33:03Z

Free Viewpoint Video for Soccer using Histogram-Based Validity Maps in Plane Sweeping

Details

Authors: Patrik Goorts, Steven Maesen, Maarten Dumont, Sammy Rogmans, Philippe Bekaert

Language: English

Year: 2014

Journal: Proceedings of the Ninth International Conference on Computer Vision Theory and Applications (VISAPP 2014)

Link: http://patrikgoorts.com/Publications/goorts2014free.pdf

Images and videos

See the projects page.

Bibtex

@inproceedings{goorts2014free,
	title={Free Viewpoint Video for Soccer using Histogram-Based Validity Maps in Plane Sweeping},
	author={Goorts, P. and Maesen, S. and Dumont, M. and Rogmans, S. and Bekaert, P.},
	booktitle={Proceedings of the Ninth International Conference on Computer Vision Theory and Applications (VISAPP 2014)},
	year={2014},
	organization={INSTICC}
}

Abstract

In this paper, we present a method to accomplish free viewpoint video for soccer scenes. This will allow the rendering of a virtual camera, such as a virtual rail camera, or a camera moving around a frozen scene. We use 7 static cameras in a wide baseline setup (10 meters apart from each other). After debayering and segmentation, a crude depth map is created using a plane sweep approach. Next, this depth map is filtered and used in a second, depth-selective plane sweep by creating validity maps per depth. The complete method employs NVIDIA CUDA and traditional GPU shaders, resulting in a fast and scalable solution. The results, using real images, show the effective removal of artifacts, yielding high quality images for a virtual camera.

Publications / Interpolation4-2013

2013-11-22T14:18:12Z

Optimization of Free Viewpoint Interpolation by Applying Adaptive Depth Plane Distributions in Plane Sweeping

Details

Authors: Patrik Goorts, Steven Maesen, Maarten Dumont, Sammy Rogmans, Philippe Bekaert

Language: English

Year: 2013

Journal: Proceedings of the Tenth International Conference on Signal Processing and Multimedia Applications (SIGMAP 2013)

Link: http://patrikgoorts.com/Publications/goorts2013optimization.pdf

Bibtex

@inproceedings{goorts2013optimization,
	title={Optimization of Free Viewpoint Interpolation by Applying Adaptive Depth Plane Distributions in Plane Sweeping},
	author={Goorts, P. and Maesen, S. and Dumont, M. and Rogmans, S. and Bekaert, P.},
	booktitle={Proceedings of the Tenth International Conference on Signal Processing and Multimedia Applications (SIGMAP 2013)},
	year={2013},
	organization={INSTICC}
}

Abstract

In this paper, we present a system to increase performance of plane sweeping for free viewpoint interpolation. Typical plane sweeping approaches incorporate a uniform depth plane distribution to investigate different depth hypotheses to generate a depth map, used in novel camera viewpoint generation. When the scene consists of a sparse number of objects, some depth hypotheses do not contain objects and can cause noise and wasted computational power. Therefore, we propose a method to adapt the plane distribution to increase the quality of the depth map around objects and to reduce computational power waste by reducing the number of planes in empty spaces in the scene. First, we generate the cumulative histogram of the previous frame in a temporal sequence of images. Next, we determine a new normalized depth for every depth plane by analyzing the cumulative histogram. Steep sections of the cumulative histogram will result in a dense local distribution of planes; a flat section will result in a sparse distribution. The results, performed on controlled and on real images, demonstrate the effectiveness of the method over a uniform distribution and allows a lower number of depth planes, and thus a more performant processing, for the same quality.