Nick Holliman

Our Cosmic Origins

Winner of the Best Computer Graphics Film Award at SD&A 2010, San Jose, CA.

Our Cosmic Origins is a sequel to Cosmic Cookery below. Created in full stereoscopic HD it visualises over six minutes the very latest simulations and observations that relate to how our galaxy formed and its place in the universe. The film was produced at the Durham Visualization Laboratory in collaboration with many researchers from Carlos Frenk's Institute for Computational Cosmology. It was first shown at the Royal Society's Summer Science Exhibition in 2009 and then again at the 350th Anniversary Exhibition in 2010 at the South Bank, London.

"We all want to know why we are here, what is special about our world, and what will happen in the future," says Carlos. "We have an emerging picture about how the formation of galaxies like our own happened just after our Universe was formed. Our computer simulations are helping us to uncover how our Universe developed into what it is today."

You can view stereoscopic 3D and 2D versions on YouTube here:
http://www.youtube.com/user/OurCosmicOrigins

Full details of the production of the film can be found in the following paper:
"Cosmic origins: experiences making a stereoscopic scientific movie", N.S. Holliman, Stereoscopic Displays and Applications XXI, Proceedings of SPIE-IS&T Electronic Imaging, SPIE Vol.7237, January 2010, San Jose, USA, ISBN: 9780819479174.

TreadWater

Treading Water

Treading Water : volume rendering from video

During Summer 2007 the Fine Artist Mara Zoltners visited the Durham Visualization Laboratory. Using her original video footage and DVL volume rendering tools an intriguing series of new images were created where time was frozen into depth.

Mara describes the work "To define new modes of expression and their use is a task that demands the questioning of relationships between the real and the virtual, and the revelation of levels of acuity of perception created by them. Working with Dr. Nick Holliman in the Durham Visualization Laboratory, I had the opportunity to investigate the imaging of movement in space over time. We created images that I call ‘volumetric temporalities’. Volumetric temporalities are images that conflate movement in space over time, representing them as three-dimensional volumes of information. In these images, the elements of space, movement, and time are represented as shaping one another rather than treated as singular events.

The series of images here, titled Treading Water (2007), are examples of our investigation. The images offer a new way of experiencing and making visible the gaps we know exists between actions but are seldom acknowledged. By investigating space and time in algorithmic three-dimensional form, possible reconsiderations of lost or hidden areas of a structure may be found. In this case, the movement of figure and water and light reflection between actions are made visible."

Full resolution images from this series can be found here.

In September 2008 our tryptic "Portrait of a wave over time" was selected by the Caladan Gallery for the exhibition "Eternal Structure".

An Expanding Shell in the Perseus B5 Molecular Cloud

Image: Nick Holliman, Durham University.

Data: Michelle Borkin, Mike Halle, Alyssa Goodman, Harvard University.

As presented at the IM2.3 workshop, this picture shows the B5 region in Perseus, a molecular cloud of gas and dust approximately 800 light years from earth, [1]. The circular feature is an expanding shell which may have been produced by a wind from an embedded young star. The data was observed using the FCRAO radio telescope in the 13CO waveband (110.201GHz) as part of the COMPLETE project, [2]. A multi-view 3D image was produced using stereoscopic volume rendering techniques at the Durham Visualization Laboratory [3] and printed in 3D using a high contrast Advanced Lenticular from Ocuity Ltd. (www.ocuity.co.uk).

[1] Michelle A. Borkin, “A COMPLETE Survey of Velocity Features in Perseus” Senior Thesis, Harvard College, May 2006.

[2] N.A. Ridge et al., "The COMPLETE Survey of Star Forming Regions: Phase 1 Data", 2006, AJ, 131, 2921.

[3] N.S. Holliman et al, "Cosmic Cookery: making a stereoscopic 3D animated movie.", SPIE Vol.6055A, January 2006.

Cosmic Cookery

2DF

Dark Matter Halo

Spiral Galaxy

Cosmic Cookery: making an animated 3D movie.

Cosmic Cookery: winner of First Prize in the 2006 vizNET Visualization Showcase

Our paper in SPIE SD&A 2006 describes experience making a short stereoscopic movie visualizing the development of structure in the universe during the 13.7 billion years from the Big Bang to the present day. Aimed at a general audience for the Royal Society's 2005 Summer Science Exhibition, the movie illustrates how the latest cosmological theories based on dark matter and dark energy are capable of producing structures as complex as spiral galaxies and allows the viewer to directly compare observations from the real universe with theoretical results. 3D is an inherent feature of the cosmology data sets and stereoscopic visualization provides a natural way to present the images to the viewer, in addition to allowing researchers to visualize these vast, complex data sets.

The presentation of the movie used passive, linearly polarized projection onto a 2m wide screen but it was also required to playback on a Sharp RD3D display and in anaglyph projection at venues without dedicated stereoscopic display equipment. Additionally lenticular prints were made from key images in the movie. We addressed the following technical challenges during the stereoscopic production process; 1) Controlling the depth presentation, 2) Editing the stereoscopic sequences, 3) Generating compressed movies in display specific formats.

You can view stereoscopic 3D and 2D versions of Cosmic Cookery on YouTube here:
http://www.youtube.com/user/OurCosmicOrigins

A short movie of Professor Carlos Frenk explaining the background to this movie can be found at:
http://www.research-tv.com/stories/technology/cosmic_cookery/bb/

Real time, smooth, multi-region depth mapping

Using non-linear stereoscopic camera models to map scene depth to stereoscopic display depth in discrete regions has the potential to create discontinuities, or kinks, in depth at region boundaries, particularly if the stereoscopic depth mapping is significantly different in adjoining regions.

Our paper in SPIE SD&A 2005 describes a new multi-region algorithm for OpenGL that overcomes this problem and can be applied to real-time graphics. It allows the scene to be divided into arbitrarily many regions and as a result intermediate regions can be introduced to smooth depth changes, avoiding visible kinks at region boundaries. The image shown has five depth regions, three main mappings and two intermediate ones for smoothing.

Mapping scene depth to stereo display depth

A continuing problem in stereoscopic imaging is how to capture depth from a scene into an image on a display so that the viewer sees a comfortable image and sees the best depth in the region of most interest. It is a particular issue for desktop 3D displays where the comfortable display depth range is limited by human factors.

Our paper in SPIE SD&A 2004 describes a new solution to this problem. The scene is split into three regions and more perceived depth is allocated to one of the regions. This allows us to vary where in the scene most perceived depth is allocated. It implements a piecewise linear stereoscopic camera model.

This is comparable to the "focus+context" methods in traditional information visualization, but we now allow the region of "focus" to move along the Z axis in and out of the perceived 3D depth range.

Testing and quality in stereoscopic imaging

This test sequence was taken to help compare different 3D displays subjective quality and technical ability to reproduce the test objects. The NPL test objects are on loan from the National Physical Laboratory in Teddington, they contain resolution & colour charts.

A single Olympus digital SLR camera was used, mounted on a Manfroto micro-adjustable slide, to take the images at 1cm intervals under fixed illumination.

Sharp Laboratories of Europe Ltd.

I was based at Sharp's European Research Laboratories for seven years where I led the software team in the 3D imaging technology group. We worked on 3D image tools for creating, storing and displaying stereo 3D images. In addition we worked on computer vision software for real time head tracking and rectifying stereo photographs for comfortable human viewing. A number of these tools have now been developed for use on the Sharp 3D display products including the Actius RD3D PC and the Sharp 3D Mobile phones.

Street Lamp

LightWork Design Ltd.

At LightWork I worked on a prototype of one of the first commerical radiosity implementations, this was capable of simulating physically accurate lighting in computer graphics environments, part of this work involved developing a novel Delaunay triangulation algorithm for illumination discontinuity meshing. This project was later developed by LightWork into the first radiosity renderer for the graphics package 3DStudio.

bearing

Leeds University and Carnegie Mellon University Ray tracing constructive solid geometry (CSG) is a CPU intensive task and my PhD and subsequent research work as an IBM Research Fellow at Leeds and Carnegie Mellon investigated how to implement parallel octree creation and parallel ray tracing of octree data structures generated from CSG models.A novel object-oriented parallel architecture was designed and implemented that used distributed caching to provide virtual shared memory on each processor. The rendering task was distributed across processors using image space task generation; i.e. sub-images were distributed to each processor. The entire control structure for the computation was also distributed so that both the computation and the control of task distribution and caching scaled with the number of processors used. Locallity was encouraged for efficient caching by using a quadtree image division to generate new tasks. The system successfully scaled to use a distributed memory parallel system consisting of over 100 processors. It was also ported to use the Nectar system, a fibre optic network of Sun workstations at Carnegie mellon.

		Our Cosmic Origins Winner of the Best Computer Graphics Film Award at SD&A 2010, San Jose, CA. Our Cosmic Origins is a sequel to Cosmic Cookery below. Created in full stereoscopic HD it visualises over six minutes the very latest simulations and observations that relate to how our galaxy formed and its place in the universe. The film was produced at the Durham Visualization Laboratory in collaboration with many researchers from Carlos Frenk's Institute for Computational Cosmology. It was first shown at the Royal Society's Summer Science Exhibition in 2009 and then again at the 350th Anniversary Exhibition in 2010 at the South Bank, London. "We all want to know why we are here, what is special about our world, and what will happen in the future," says Carlos. "We have an emerging picture about how the formation of galaxies like our own happened just after our Universe was formed. Our computer simulations are helping us to uncover how our Universe developed into what it is today." You can view stereoscopic 3D and 2D versions on YouTube here: http://www.youtube.com/user/OurCosmicOrigins Full details of the production of the film can be found in the following paper: "Cosmic origins: experiences making a stereoscopic scientific movie", N.S. Holliman, Stereoscopic Displays and Applications XXI, Proceedings of SPIE-IS&T Electronic Imaging, SPIE Vol.7237, January 2010, San Jose, USA, ISBN: 9780819479174.
		Treading Water : volume rendering from video During Summer 2007 the Fine Artist Mara Zoltners visited the Durham Visualization Laboratory. Using her original video footage and DVL volume rendering tools an intriguing series of new images were created where time was frozen into depth. Mara describes the work "To define new modes of expression and their use is a task that demands the questioning of relationships between the real and the virtual, and the revelation of levels of acuity of perception created by them. Working with Dr. Nick Holliman in the Durham Visualization Laboratory, I had the opportunity to investigate the imaging of movement in space over time. We created images that I call ‘volumetric temporalities’. Volumetric temporalities are images that conflate movement in space over time, representing them as three-dimensional volumes of information. In these images, the elements of space, movement, and time are represented as shaping one another rather than treated as singular events. The series of images here, titled Treading Water (2007), are examples of our investigation. The images offer a new way of experiencing and making visible the gaps we know exists between actions but are seldom acknowledged. By investigating space and time in algorithmic three-dimensional form, possible reconsiderations of lost or hidden areas of a structure may be found. In this case, the movement of figure and water and light reflection between actions are made visible." Full resolution images from this series can be found here. In September 2008 our tryptic "Portrait of a wave over time" was selected by the Caladan Gallery for the exhibition "Eternal Structure".
		An Expanding Shell in the Perseus B5 Molecular Cloud Image: Nick Holliman, Durham University. Data: Michelle Borkin, Mike Halle, Alyssa Goodman, Harvard University. As presented at the IM2.3 workshop, this picture shows the B5 region in Perseus, a molecular cloud of gas and dust approximately 800 light years from earth, [1]. The circular feature is an expanding shell which may have been produced by a wind from an embedded young star. The data was observed using the FCRAO radio telescope in the 13CO waveband (110.201GHz) as part of the COMPLETE project, [2]. A multi-view 3D image was produced using stereoscopic volume rendering techniques at the Durham Visualization Laboratory [3] and printed in 3D using a high contrast Advanced Lenticular from Ocuity Ltd. (www.ocuity.co.uk). [1] Michelle A. Borkin, “A COMPLETE Survey of Velocity Features in Perseus” Senior Thesis, Harvard College, May 2006. [2] N.A. Ridge et al., "The COMPLETE Survey of Star Forming Regions: Phase 1 Data", 2006, AJ, 131, 2921. [3] N.S. Holliman et al, "Cosmic Cookery: making a stereoscopic 3D animated movie.", SPIE Vol.6055A, January 2006.
		Cosmic Cookery: making an animated 3D movie. Cosmic Cookery: winner of First Prize in the 2006 vizNET Visualization Showcase Our paper in SPIE SD&A 2006 describes experience making a short stereoscopic movie visualizing the development of structure in the universe during the 13.7 billion years from the Big Bang to the present day. Aimed at a general audience for the Royal Society's 2005 Summer Science Exhibition, the movie illustrates how the latest cosmological theories based on dark matter and dark energy are capable of producing structures as complex as spiral galaxies and allows the viewer to directly compare observations from the real universe with theoretical results. 3D is an inherent feature of the cosmology data sets and stereoscopic visualization provides a natural way to present the images to the viewer, in addition to allowing researchers to visualize these vast, complex data sets. The presentation of the movie used passive, linearly polarized projection onto a 2m wide screen but it was also required to playback on a Sharp RD3D display and in anaglyph projection at venues without dedicated stereoscopic display equipment. Additionally lenticular prints were made from key images in the movie. We addressed the following technical challenges during the stereoscopic production process; 1) Controlling the depth presentation, 2) Editing the stereoscopic sequences, 3) Generating compressed movies in display specific formats. You can view stereoscopic 3D and 2D versions of Cosmic Cookery on YouTube here: http://www.youtube.com/user/OurCosmicOrigins A short movie of Professor Carlos Frenk explaining the background to this movie can be found at: http://www.research-tv.com/stories/technology/cosmic_cookery/bb/
		Real time, smooth, multi-region depth mapping Using non-linear stereoscopic camera models to map scene depth to stereoscopic display depth in discrete regions has the potential to create discontinuities, or kinks, in depth at region boundaries, particularly if the stereoscopic depth mapping is significantly different in adjoining regions. Our paper in SPIE SD&A 2005 describes a new multi-region algorithm for OpenGL that overcomes this problem and can be applied to real-time graphics. It allows the scene to be divided into arbitrarily many regions and as a result intermediate regions can be introduced to smooth depth changes, avoiding visible kinks at region boundaries. The image shown has five depth regions, three main mappings and two intermediate ones for smoothing.
		Mapping scene depth to stereo display depth A continuing problem in stereoscopic imaging is how to capture depth from a scene into an image on a display so that the viewer sees a comfortable image and sees the best depth in the region of most interest. It is a particular issue for desktop 3D displays where the comfortable display depth range is limited by human factors. Our paper in SPIE SD&A 2004 describes a new solution to this problem. The scene is split into three regions and more perceived depth is allocated to one of the regions. This allows us to vary where in the scene most perceived depth is allocated. It implements a piecewise linear stereoscopic camera model. This is comparable to the "focus+context" methods in traditional information visualization, but we now allow the region of "focus" to move along the Z axis in and out of the perceived 3D depth range.
		Testing and quality in stereoscopic imaging This test sequence was taken to help compare different 3D displays subjective quality and technical ability to reproduce the test objects. The NPL test objects are on loan from the National Physical Laboratory in Teddington, they contain resolution & colour charts. A single Olympus digital SLR camera was used, mounted on a Manfroto micro-adjustable slide, to take the images at 1cm intervals under fixed illumination.
		Sharp Laboratories of Europe Ltd. I was based at Sharp's European Research Laboratories for seven years where I led the software team in the 3D imaging technology group. We worked on 3D image tools for creating, storing and displaying stereo 3D images. In addition we worked on computer vision software for real time head tracking and rectifying stereo photographs for comfortable human viewing. A number of these tools have now been developed for use on the Sharp 3D display products including the Actius RD3D PC and the Sharp 3D Mobile phones.
		LightWork Design Ltd. At LightWork I worked on a prototype of one of the first commerical radiosity implementations, this was capable of simulating physically accurate lighting in computer graphics environments, part of this work involved developing a novel Delaunay triangulation algorithm for illumination discontinuity meshing. This project was later developed by LightWork into the first radiosity renderer for the graphics package 3DStudio.
		Leeds University and Carnegie Mellon University Ray tracing constructive solid geometry (CSG) is a CPU intensive task and my PhD and subsequent research work as an IBM Research Fellow at Leeds and Carnegie Mellon investigated how to implement parallel octree creation and parallel ray tracing of octree data structures generated from CSG models.A novel object-oriented parallel architecture was designed and implemented that used distributed caching to provide virtual shared memory on each processor. The rendering task was distributed across processors using image space task generation; i.e. sub-images were distributed to each processor. The entire control structure for the computation was also distributed so that both the computation and the control of task distribution and caching scaled with the number of processors used. Locallity was encouraged for efficient caching by using a quadtree image division to generate new tasks. The system successfully scaled to use a distributed memory parallel system consisting of over 100 processors. It was also ported to use the Nectar system, a fibre optic network of Sun workstations at Carnegie mellon.

Image Gallery