The Early Buzz
Brad, I just took a look at the Cirri
Scans on Sketchfab and they look great! Definitely professional grade. You can
quote that. :: Alban Denoyel Co-founder & CEO sketchfab.com @sketchfab
- WOW -- that's incredible! What can I do to help? :: Lightwave 3D Group Support ( LightWave is responsible for more artists winning Emmy Awards than any other 3D application).
At Allegorithmic we’re all about
making the life of 3D content creators better and more exciting. CIRRI’s system
is exactly about that, and the quality you can get out of it seems amazing.
We’re excited and proud to be part of this project. :: Sebastien Deguy, CEO Allegorithmic
- There's some good looking models there.. The import was simple, as expected, and the model was ready to sculpt and paint right away. :: Simon de Rivaz, Curvy 3D
I have never seen a product that
could provide us with such a unique solution and tool for heritage stone
work. BRAVO! :: Marc Langlois,Head of Gem Campbell Eastern
- The software from the preview looks absolutely impressive! Taking nice 3D model from set of photos is just a dream. :: Andrew Shpagin, 3D-Coat
These 3D models look great in AC3D. Cirri will be an excellent resource for our users. :: Andy Colebourne, Inivis AC3D
The Cirri technology team has been at the forefront of 3D scanning developments for the last 30 years
Do you know that Cirri co-founder and chief of technology, Gary Robertson, developed the world's most accurate scanner in the early 1980s, with an accuracy of 1 part in 300,000. Then in 1986 developed the worlds first scanner that used photographs to generate the 3d model almost 8 years before the term photogrammetry (using photos to make a 3d model) was even coined. His 67 published and peer reviewed papers have been at the forefront of almost all developments in 3D for over 30 years. That's the pedigree of Cirri.
Interactive 3D models: Click on any image with a box in a circle to open a 3D model created by Cirri. When the model opens use your mouse and scroll wheel to interact with it. Click on any image with a Start arrow to watch a video (note some models are large and may take a few extra moments to load, depending on your internet speed)
Change your Perspective - Change your World
Two photos of the gargoyle used in the scan. The whole model seen below was captured from 3 pairs of photos.
A scan creates digital models from the real world. This scan was captured by talking two photos of this gargoyle from the sidewalk and turned into a 3d model. How is that possible? Read on. We also have an FAQ section at the bottom of the campaign that answers specific questions we've received. If you're unclear just post a question.
Above: The gargolye scan loaded into Lightwave (courtesy of Lightwave)
What is our Project?
Our project is to create a new 3d scanner based on the smarts underlying the worlds best professional scanning technology. Our project will consolidate the hundreds of parameters used in that technology into a step-by-step, easy to understand process appreciated by both professionals and amateurs to capture the real world in 3D as it actually is. This project will generate two new software based scanning products called Cirri and Cirri Lite, that transform digital photos into 3d models. Cirri provides scale, ground based coordinates and accurate measurements. The Lite version captures the model but without scale and coordinates. Both products will be supported by online training tutorials and teaching manuals - Help is a click away. There is no compromise to quality. Cirri is no gimmick. You will create point clouds of intense beauty, utility or both.
Every scan and resulting model shown on the Cirri 3d Scanner Indiegogo campaign, and our website www.cirri.co, were captured with parameter sets that will be automated in the Cirri user interface. The scan results you see in the campaign are the scan results you’ll get from Cirri. (scroll down to: It sounds too good to be true, for more about this and our Game Plan). More questions? There is an FAQ sections at the very end of the presentation.
It's all about the Quality
Cirri is a startup formed by the 3D technology pioneers going back over 30 years. Most of the 3D scanning technology is based on the Cirri team's developments. Cirri's goal is to make 3D as easy as 2D. Our strategy is based on delivering high quality, accurate scanning results. We know that you've got to start with quality. Low quality scans = frustration. High quality scans = endless possibilities. We can deliver that.
A uniform, dense point cloud captures the essential detail needed to use the model in other applications
We’re launching our first Crowd funding campaign to bring you professional grade 3D scanning capabilities. If you don't know what that means, consider that professional grade 3d scanning tools can cost from $15,000 to $30,000 for software, and rising to over $100,000 for some hardware scanners. Our software provides the same professional quality results at price points that the rest of us can afford.
What makes a Cirri scanner Unique?
Cirri's technology is better than anything in the market today, including what's used by Microsoft, Apple, Autodesk, 3D Systems or any other. We have lead 3D technology developments for over 30 years. We continue to push all boundaries.
Unlike any other scanning devices in the market today that require additional hardware, suffer from poor resolution outside of a narrow parameter, limited range of subject or object size, proximity limitations, a large number of photos of the object, poor quality of scan data and no accurate scale; Cirri combines professional quality scanning results, accurate, to scale, high resolutions scanning, almost unlimited range of object size or geographic feature, minimal number of photos (2 per visual face) all at a remarkably low price. Cirri is developed and supported by a team of true technological pioneers and visionaries in the realm of 3d scanning. We know what is needed.
A cathedral scanned, measured and modeled.
How accurate and
how much detail?
Cirri scanners can with the proper setup achieve accuracy of less than one third the thickness of a human hair with sufficient resolution to capture the shape of the relief on a coin.
Cirri products are based on 3D tools used today by major video game and movie studios, as well as universities and a cross section of Fortune 500 companies. The products and their underlying technology were developed by Cirri’s chief technology advisor and partner, Gary Robertson, a pioneer in the field of close range photogrammetry. Gary’s ShapeCapture product suite enables professionals to create point clouds of the highest quality. Cirri uses a modified scanning engine from ShapeScan to drive results.
Zooming in you can see the uniformity of the point cloud
Point Clouds Rule
Point Clouds form the foundation of all 3D scanning. They define shape and detail and they look, somewhat, like Cirrus clouds which is how we chose our name. If you Google point cloud; passive stereo imaging, photogrammetry, metrology, normal mapping, and cirrus clouds, you’ll get the gist.
It's Easier than it Looks
A face captured for a movie - the points are used by animators, not for the Cirri. The full 3d head to the right was captured separately to the actor's face and used as part of a movie's special effects. It allows the actors face to replace that of the stunt person during action sequences (pop goes that bubble).
Creating professional grade three dimensional models can look and feel a little bit daunting, We get it that this might be all new to you - I'm here to tell you that you don't need to be a techy or some kind of computer programmer to create everything we're showing you. You only need a desire to create quality three dimensional models of the real world - of your world. That's all you need. We'll give you the tools and training to make it so. So please read on to learn more about what you can do and how you can help us get you there.
From photos to 3d model
Who is Cirri for?
If you've read this far it's for you!
It would help if you knew a little bit about 3d because it is a bit more involved than clicking a camera shutter. But have no fear our mission is to make it easy. That's why we require funding to create a user interface that walks you through the process. Cirri is for a professional or anyone demanding professional results with all the demands of perfection. Yet we're going to make it a tool useful for people (like you) who might be pros or dedicated lovers of 3d or the super interested who want to capture your world without needing a manual. Our point is we intend to make it easy. The long term goal we're striving for is to make 3d modeling of the real world as easy as, well, taking a photo. But that's in the future. Right now we know there's a little more work. So Cirri today is for:
An animator, artist, modeler, heritage stone works fabricator, designer, studio manager, landscaper, paleontologist, gamer, photographer, sign maker, theme fabricator, entrepreneur, product designer, marketer, producer, fabricator, 3d printer, cnc fabricator or the merely curious. Cirri brings fun and utility dimensionalizing our world.
Okay more app potentials: You capture 3d models of loved ones to be treasured forevermore, and don't forget the dog. You're a geologist, historian, an architect, archaeologist, anthropologist, a CSI investigator, a prospector, a developer or a cartographer. You're an athlete finding the edge; a cinematographer enhancing realism or a granite contractor requiring precise dimensions of a building facade - the list is almost endless. Cirri can take you there on a foundation of quality and simplicity.
Capture highly detailed geometry from the sidewalk
From Point Cloud to anywhere...
There are literally thousands of uses for a quality, accurate 3D scan. To that end, we’ll map out paths to go from the point cloud into an array of applications. Paths to video gaming, movie animation, 3D printing, CNC cutting, metrology, reverse engineering, mapping, cataloging, 3D image viewing. We will keep adding new paths and give you the ability to contribute tutorials and post helpful tips and techniques on our web forums and user groups.
It sounds too good to be true,
and as the saying goes...
...but in this case it is true. Cirri is a disruptive product. We know it works because at its working core is the same software as its more expensive parent that is in use every day around the world. The parent product targets a different crowd that require huge computing horsepower. For Cirri we limit calculations to around one million measured points and provide high, but not subpixel, accuracy. We recognize the world is fast evolving in 3d where desktop 3d printers, dimensionalized websites and immersive and personalized video gaming is fast approaching. 3d is stepping out from the realm of the professional. Not to be boastful but we know our underlying scanning engine, algorithms and techniques are far ahead of the other approaches.
To gain traction 3d has to have a foundation of quality. The other approaches that sense 3d or require a huge number of photos as input, in the end, lead to disappointment and disillusionment.
So we created Cirri, We priced it fairly and we will streamline the user interface so you don’t need to be a professional to use it – though we hope professionals too will love it.
The game plan
Automate the parameters based on a few simple input steps.
1. Tell Cirri about your camera by inserting a calibration file (which we walk you through how to make), and choose how many megapixels in the photos. Once this is setup it can be used over and over, or you could setup several cameras then choose the camera like you choose a printer.
2. For each scan session you take photos of the object you want modelled.
a. Enter the approximate distance you are from the object you’re photographing,
b. Choose one of (6) images that most closely represents the scan ie: trinket, , face, body, wall carving, building, mountain. These choices automatically establish the initial scan parameters.
c. Enter a scale which makes the measurements accurate in real world coordinates
d. Click to make the model. After partial number crunching Cirri gives you three preview results.
e. You choose what looks like the best result.
f. The finished 3d scan emerges
g. There may be more tweaking through slide bars and/or another round of preview images – this will be determined by our beta testers.
h. The finish result is a high resolution model created from the photos (including full 360 degrees. As well the photo images are blended and matched to the surface so that it appears as if photographed from one image.
i. We’ll have send-direct-to-other-software/utility options, or the 3d model can just be saved using a number of common 3d file formats.
The video below shows the results from scanning a vehicle test track from photos taken from a boom lift 21 feet above the road.
A real world example
If you want to see more please scroll down to the application area further on in this presentation.Here's an example of the steps to go from - an original carving - to Cirri scan - to CNC toolpath software - to CNC machine - to finished replicated carving.
Similar to 3D printing, a CNC machine benefits from capturing as much detail as possible in the scan. The limit on a CNC is the diameter of the cutting bit used to route into the material. Fine variations or knife style cuts don't reproduce on the cnc. The exception is if you expand the original scan significantly larger then those details may be utilized. This is common practice in the Theming industry.
Original Carving - take a couple of photos and load them into the Cirri software
Made into a digital model by Cirri Lite
Import the model in a CAD/CAM software for CNC machining. In this example I've only chosen the middle section of the carving. In the CNC toolpath software I re-sized the model to fit my material. In this case my material is smaller than the original carving. I also smoothed out some detail as the bit diameter will not pick it up in the carving and the machine runs quicker over a smoother surface.
Carving inserted into CAD/CAM software program (Aspire, by Vectric, U.K.)
View Simulation of the machine carving to make sure it looks right. Finished CNC simulation in hardwood looks good
Run toolpath on a CNC machine. CNC on the finishing pass of Carving (Intelicam, CNC router)
Here's a video showing the finish pass using 3mm ballnose router bit.
The finished CNC carving. Additional details can be carved into the wood by hand before final finishing. Original - to Cirri - to CNC - to new carving
· Products/tools used in this CNC application example: Uncalibrated digital camera; Cirri scanner scanner; Aspire CAD/CAM software; Intelicam CNC router.
Museum Totem Pole
What if I'm an application developer?
Our industry standard file formats provide a vast opportunity for application developers and partners to package up your own solutions. We've established a special reward category just for you! What that category means is that you'll have an open line to us while we're evolving the user interface, and you're developing solutions that leverage our quality point cloud.
We encourage all developers to create your own apps that adhere to the overriding parameters of quality, intuitive user interface and an abundance of tutorials. We want users of our tools to always knows how to get to there from here. You as developers will use the point cloud in ways we might never have imagined. We will work together to create an online space to share or sell your applications. We’re itching to go.
Can I OEM Cirri in my own product?
Quick answer is yes. OEM arrangements are a negotiated arrangement often involving non-recurring development costs and minimum quantities. We are happy to discuss these arrangements with you. We've opened up a special reward section for companies and individuals called OEM Partners. Our engineering and development resources are limited, so to be fair we allocate resources to OEM partners on a first come basis.
Subsea pipeline photo by deepwater ROV
3d model of the pipeline with precision measurements
The first product from Cirri is based on our version of a passive image stereo match technology. We call it Cirri after the company name and the cloud formation it resembles. It’s a fantastic tool that lets you capture textured objects as diverse as faces, to sculptures, to mountain sides - all using a standard digital camera or smart phone. We are starting our Crowd funding campaign with pre development completion pricing for Cirri from $479 up to $599, depending on how soon you sign up. Cirri allows you to scan to very high accuracy and includes full colour textures and ground based measurement coordinates. Those coordinates apply an accurate scale to the dimensions which is important for many applications.
Cirri allows you to accurately scan a mountain side
Cliffside - scanned and measured
We will also release a reduced feature scanner called Cirri Lite. This will still produce the same high quality point cloud as Cirri but will not have features like scale or ground base coordinates built into it. The bonus is we can offer it at a much lower price. Our theory is if you don't need that stuff, why pay for it. We have student pricing available - see the rewards section for how that works.
The objective of the campaign funds are to make the creation of professional grade point clouds accessible to even casual users. To do this we will create simplified user interfaces for Cirri and Cirri Lite, training videos, manuals and map out directions to go from the point cloud to an array of end user applications.
Our goal is to get Cirri into the hands of early adopters, product development partners and the curious like you as soon as possible to help us shape product development leading up to the first release version. We're passionate about 3D. We know that starting from quality changes the purpose of a scan from a fun, cool gimmicky thing to a useful tool providing extraordinary utility for a range of applications. We want partners to create products that use the quality scan as a foundation to their own great products to the benefit of everyone. Indiegogo gives us access to an extended community that helps us shape the direction and functionality of Cirri.
What Digital Cameras will work?
Cirri is not specific to any particular digital camera. Generally the better the camera the better the result. A two mega pixel camera will provide the minimum resolution. Clear focus is important. The quality of the camera's sensor is also important. The digital camera built into the iphone 5S or 5C passes the mark. Here are a few of the hundreds of cameras that will produce great scans.
Many different digital cameras will work fine.
Do I need to calibrate the camera?
Perhaps. It depends on the camera. The scan of the Indian wood carving was created with an uncalibrated Sony NEX camera and it turned out fine. Some digital camera sensors have significant distortion, which affects cirri's ability to create a model. Fortunately these are constants and can be corrected through calibration. It is a simple process and once that calibration is known by cirri you don't have to worry about it.
Minimum Computer Spec
Cirri currently functions best on a minimum Windows 7, 64-bit operating system with 8 Gb RAM and a video card.
Don't despair if you don't have that computer configuration. Cirri can function on a lesser spec, but you will have some limitations mostly based around resolution, which affects file size. So you can use a computer with only 4 gig RAM. A lot of us use the 32-bit Windows version even though 64 bit is available at the same price. 64-bit did have its issues but since Windows 7 those are mostly resolved and you do get far more processing activity. We have the capability to create a 32-bit compatible Cirri and will do that if required, so no worries.
We are not Mac compatible but again that is because we have not had a need to be. We do have the capability to become Mac compatible if there is sufficient demand. So please let us know.
Like Mac compatibility we also have the capability to have CIRRI available as a Cloud based application. Lots of good reasons exist for that to happen, we just need (enough of) you to tell us that is what you want.
Cirri Scanning Limitations
Cirri works best for objects with texture. Texture is best described as non-uniform surfaces with lots of detail. Look around you. Most items within eyesight have lots of texture and produce great scans. But not all. Objects like a boat hull or a plain white wall do not scan well with Cirri. There are techniques to help Cirri take measurements but they require additional steps.
An application starts from the Cirri scan, then by combining different components or techniques an application takes you to a place of new utility or fun. The possibilities are endless. We've already posted a work flow for CNC applications (see above). This also applies to many applications including set design, theming, architectural features. So let's take a look at some others:
3D modeling for Video Games, Movies and Animation
A wide array of software is employed to transform a Cirri scan into elements used in video games, animation, movies and other digital art and media. If you're in that world then you are likely already familiar with what you need and what to do. If you're new to this space and want to learn more we will be supporting user forums and support groups to help you go from a cirri scan to whatever you need to do.
In this section we'll show you screen captures of some of the most popular software that can import a cirri scan directly. Software programs are listed in no particular order. We'll keep adding as we get more screen captures. Please contact firstname.lastname@example.org.
(image below courtesy of Lightwave 3D)
WOW -- that's incredible! What can I do to help? :: Lightwave 3D Group Support
(2 images below courtesy of Curvy 3D)
Thanks for the Obj... I lit the model with ambient occlusion and shadows to enhance the subtle shading included in the scan. best regards, Simon
AC3D by Inivis
(2 images below courtesy of Inivis)
These 3D models look great in AC3D. Cirri will be an excellent resource for our users, Andy Colebourne
(meshlab is a .org with open source software)
(2 images below courtesy of 3D-Coat)
(image below courtesy of Cheetah3d.de)
Posting 3d images online
How a 3d scan is changed so it can be posted online (Explainer)
As you might have imagined quality 3d scans have lots and lots of data in the form of three dimensional topology with a photo wrapped overtop. The files are large. Too large to upload then download in any reasonable time frame. These files are referred to as being heavy. The magic with posting online is to make a heavy file light while retaining the visual detail captured from a high resolution scan. Next is a heavy file made up of 950,000 triangles (Sketchfab does a great job of opening it, but it's still pretty slow). I've zoomed into a portion of the corbel so you can view the triangles. Click on the full model to view in Sketchfab.
3D model above has 950,000 triangles defining the shape. The same 3D model below has been decimated with the normals burned into the texture by our friends at Allegorithmic. It now has only 1,500 triangles. If this were used in a video game that would be further reduced to 10 or 12 triangles, just enough to show dimension but the texture still captures the visual appearance of 950,000 triangles. This is why it's so important to start with quality - high resolution - high accuracy.
Here's a photo of the object showing the outline of each of the 1500 triangles. Following it is the same image as an interactive model.
To do this you want to retain the topographical detail (wrinkles and bumps) inherent in a heavy file yet get rid of file size. And that's just what is done. It's called burning a normal map (or height map). To understand what a normal map is, you need to separate a wrinkled surface into two parts. The top part is a photograph. The second part is the underlying physical shape of that wrinkled surface. A normal is kind of a third element to all of this. It looks like an arrow and is positioned perpendicular to the direction of each point on the underlying shape.
A normal map
A normal determines the intensity of the light reflected on each point on a wrinkle in relation to the direction of the light. A normal pointing at a light source glows intensely. A normal angled away from the same light source appears in shadow. As the sun crosses the sky the normals visually reveal the shape of the wrinkles on the surface of the earth, like a canyon or a creek bed. Normals also reveal the details in a coin or the bush strokes on a painting.
So the magic occurs when you"burn" those normals and forever lock the direction of each arrow without dependency on the underlying shape. The photo itself now holds the normal. The underlying shape is redundant and most of it along with most of the weight of the file, can be removed. The detail is retained the shape is mostly gone so the file very much ighter. Light enough to post online.
Here's another example of the before and after borrowed from Wikipedia: 4 million points reduced to 500 triangles - now you can post it online
This illustrates why creating a professional quality 3d scan is critical to being useful regardless of the minimum allowed file size for eventual output. Cirri and normal maps are best friends.
Some 3D printers fabricate only a shape while others print the shape with a full colour surface (texture). The resolution of the print can vary considerably from one model to the next. Cirri can produce all of that. In 3D printing you don't want to reduce the resolution like you do in a video game. You need the 3D printer to deposit material in every nook and cranny from the scan so the output represents the real object. For uses in reverse engineering or product prototyping accuracy of scale is critically important.
3D Printed mono-color heads
Color 3D printed head
The issue in 3D printing is that the scan must not have holes along a continuous surface. This is referred to as being watertight. A 3D printer doesn't like blank areas and a scanner can only measure what it can see. So there are often inherent blank areas or "confused" gaps that prevent a watertight object. We have simple techniques to deal with that:
Finding and patching holes to make a watertight object
I was explaining Cirri's unique ability to capture 3 dimensional swaths of the earth's topology to a friend who is an avid mountain climber. She pointed out that Google Earth already provides terrain dimension information and wondered how Cirri can improve upon that. Google Earth is an excellent resource but their sampling rates can vary between about 15 feet up to over 100 feet.
Google earth terrain - excellent reference but needs detail
Cirri compliments Google Earth by overlaying higher resolution, more accurate dimensions and visual detail within Google Earth. For instance Cirri can measure the precise pitch of a rock face to within feet or inches; the distance from one face to the next; the angle of a ski slope to negotiate a turn from one gate to another.
Here's how Cirri works within Google Earth.
Using my GPS I can enter real earth coordinates at a few points on my Cirri scan. Cirri embeds that information and outputs the scan in a Google Earth friendly format. Now the scan opens up directly in Google Earth and my scan (of the rock face) shows the high resolution and accuracy I desired yet still having the perspective of the rest of Google Earth interface to use as I please. It's an extraordinary tool.
- posting to Google Earth
Terrain Mapping a Golf Course
Measuring and modeling the subtle undulations and elevations of your golf course fairways and greens is within your reach.
Golf green topology
rendered 3d golf hole
Cirri has some novel approaches to help you do just that including mounting a camera on a remote controlled quadcopter.
GoPro mounted to remote controlled quadcopter
- historical records and curating
We showed you a cathedral earlier. Here's a 3d model of an historic carving. It's a large file so may take a few more seconds to open.
Reverse Engineering, Inspection
Important note. Cirri does have scanning limitations (see Limitations section above), and these limitations can be more pronounced in man-made objects that are often the focus of reverse engineering and inspection. Please understand these limitations or contact email@example.com for more information.
An accurate point cloud is an invaluable component of part inspection and reverse engineering. The basic idea of how it works is that a manufactured part that originates from a 3 dimensional CAD design can be inspected using a 3d scanner to measure what was manufactured. The point cloud can then be overlayed on the original CAD model. Deviations from the original model to the real part are calculated and determined if the part is within acceptable tolerances. This requires additional software to perform.
Here's how that usually looks:
The scan data is combined with the original CAD
Prosthetic and medical devices
Cirri captures the complex shape of the human body which makes it a useful tool for designing a prosthetic, as illustrated in the following examples (note these prosthetics did not originate from a Cirri scan and are for illustration purposes):
Underlying scan geometry of the hand was used to create prosthetic
A precise scan of the limb is used to model an optimal fit.
Risks and challenges
The risks with completing
our Indiegogo campaign funded project are minimal. The key risk is mitigated
because the underlying scanning software engine already exists and in use
everyday. One hurdle we face is the work required to create a streamlined user
interface for Cirri and Cirri Lite. We are confident that we can do that
owing to our decades of experience creating software products. In addition part
of our development strategy is to seek feedback from the development partners
and beta testing groups that joined us from this campaign.
We are not relying on any one of us as developers and managers but have split the work among several people so if something should happen we are not left without an alternative.
There is a risk with timeline. However we have managed dozens of software and hardware development projects over careers that span decades and feel with confidence that the dates in our production plan are achievable.
So we think we're in a
pretty good position to deliver good on our word.
This is what we do for our livelihood and it would certainly not be in any of our interests to deliver on less than our word.
Brad Moore brings an extensive management background to the CIRRI team. He has
been an active participant in the development and marketing of 3d measurement
and visualization technologies, identified and pursued technical developments
and market conditions leading to the formation of CIRRI.
Brad was twice nominated for the Ernst & Young Canadian Entrepreneur of the Year award while CEO of a Western Canadian based company that developed, manufactured and sold industrial cnc machinery and computer based manufacturing solutions with installations in 60 countries worldwide. He has hardened experience to achieve profitability and stability with both technology startups and established manufacturing companies. His companies have been the recipient of many awards including the Calgary Export Achievement award and the Coolest Business in Alberta award. Brad is a McMaster
University Letterman. He has lead companies to eight figure valuations and facilitated business owner exits leading to multi-million dollar payouts.
Chief Technology Advisor
Gary Robertson has worked solely in the field of close range photogrammetry for over twenty-five years. He has undertaken photogrammetric consulting and research and development in all areas of science and engineering, including machine vision systems and microdensitometry.
He is a member of the American Society of Photogrammetry and is a board certified registered photogrammetrist (ASP). He has authored over sixty five peer reviewed technical papers in areas of aerospace, forensics, geotechnical and civil engineering, robotics and medical applications. He has presented papers internationally on five continents, and has been a contributing author for several books relating to Photogrammetry.
He has worked internationally on accident and forensic applications. He has developed procedures for photogrammetric applications in the accident and forensic field that are being used in Australia, Canada, United States and the United Kingdom.
Gary Robertson investigative cases have been shown on TV shows such as Forensic Files and Discovery Channel.
Gary has worked in the motion picture industry with 3D modeling for visual effects and development of process for 3D face replacement, or digital doubles with pending patents. He is also a co-inventor of a hardware, software system for Spatial Data Capture for 3D Stereo Imaging, This system approach greatly simplifies the addition of computer generated graphics, animation and other special effects in cinemagraphic post-processing.
Gary Robertson holds training
programs in Close Range Photogrammetry and high resolution imaging with
international participation. He has been a invited speaker to several
international organizations such as the ISPRS(International Society for
Photogrammetry and Remote Sensing) IAI(International Association of
Identification) In addition to lecturing at the Royal Canadian Mounted Police
training facility and the Canadian
He received B.Sc.(Hon.) in Mathematics and Physics from University of Hong Kong in 1970, and Graduate Diploma in Computer Studies (first in class) from University of Melbourne in 1990. From 1990 to 1995 he did postgraduate research in mathematical programming and computational geometry. He published several papers in operations research and signal/image processing.
He has been in charge of over 40 research projects in industries, universities and research organizations. Two third of his projects are in 3D imaging, image processing and 3D modeling, using mathematics, artificial intelligence, optimization and computing techniques. He authored several patents and pending patents. He has developed a large number of innovative algorithms in high-density stereo matching with/without structured light, image rectification & registration, mesh processing, object recognition and nurbs surface modeling, as well as tie-point alignment, mesh stitching and texture mapping for a large number of images. He also provided novel solutions to custom 3D imaging problems for different medical and engineering applications.
Patrick is an accomplished and award-winning entrepreneur with both domestic and international experience. As a business leader and builder he has established and been responsible for growing companies, across the spectrum, from start-ups to eight-figure revenue. An innovative and strategic thinker, Patrick is a results orientated, decisive leader and has a proven ability to quickly analyze and assess key business drivers and metrics and develop, implement and execute on strategies and models to maximize stakeholder value.
Patrick has participated as an angel investor and has secured multiple levels and stages of financing and built all collateral and materials necessary to obtain funding. Implemented systems for cashflow monitoring and ratio maintenance.
Here are answers to questions we've been getting.
Can you scan hair?
Yes cirri can scan hair, you can see an example of that on our campaign site. However the hair does end up looking somewhat like a helmet, but for 3d printing, say, you need that solid structure anyway. If a human face is being used in a movie or videogame the hair is usually pulled back so that an animated hair can be re inserted during animation or special effects sequences.
What happens to the areas that are not seen in the photos - does Cirri leave a hole?
If you look at the gargoyle or if you look at the image of the actor's face, you can see the pair of photos does not cover per say the inside of the nose on the face. For example you see the left from the left but not the right etc. That does not leave a hole. Cirri fits and fills that occlusion and models it to the surface. Cirri does not need to have a matched pair for all surfaces. It will use coordinate geometry in sync with stereo match photogrammetry. It is one of the methods we employ to reduce the number of photos.
How many photos do you need to create a scale 3d model?
Can I scan an airplane?
An airplane starts to
fall within the limitations as the software, which requires texture
(variability of the surface), and this is not typical of an airplane
What differentiates you from 123D?
The 123D algorithm is based on an open source freeware and requires a large number of photos and hence significant computing power to create anything. Cirri requires 2 photos per viewable face and can fill in the rest. In addition our point cloud generation is more even/regular hence more usable (a prerequisite part of making it easy). The Cirri output has a very clean, dense point cloud which you control - quality is key. Cirri has ground based coordinates. Lastly you can own cirri and the 3d models you create without uploading anything.
Can you do 3d capture from video?
Yes provided each video image meets our minimum recommended spec (2Mp).
How do I know I'll be chosen to be a beta tester?
The first 100 people who sign up as beta testers are accepted. After that it is closed. Please go to the Perks contribute section (right side column) to sign up.
Which CNC machines will your software support?
Cirri can output an stl file format into a CNC machine's CADCAM software. The CADCAM software is what drives your machine.
Stl (stereo lithography) is the most common 3d mesh type of file used in CNC machining. That is the file output I used on the example of carving the Indian carving scan in this campaign. For that example I used Vectric's Aspire.
Many of the most common CADCAM software can open, manipulate and toolpath a 3d mesh based file. Popular CADCAM software that comes to mind are ArtCam, Type 3, MasterCam, RhinoCam, MeshCam, FreeMill, Deskproto, PowerMill and many more that work, Rhino3D has an excellent resource page on their website. All of those softwares can load a 3d mesh model, toolpath, generate an NC code and drive your machine. This means that Cirri (indirectly) supports any CNC machine that accepts G-code from these softwares (which is pretty much all CNCs).