cgspeed

A Carrara 8 test

2010-11-25T20:44:00.001-08:00

I finally took advantage of the deep-discount offer on Carrara 8 Pro provided by 3D World magazine issue 133, which shipped with a copy of Carrara 7 Pro on the enclosed DVD plus an upgrade-to-8 offer. This animation is a variant of previous work, however here the rendering is done in Carrara, which also provides the sky and ground plane. Daz Studio could also of course provide the ground and the drop shadows, though it doesn't do skies. The embedded video below isn't high-def, so for the better (720p) version, go view it directly on Vimeo and click the player's expand-to-fullscreen button.

The workflow here was fairly simple:
- Save the previously-created (and animated) Daz Studio version using the "Daz Collada" format, the Daz variant of Collada designed for moving scenes from one Daz program to another.
- Load into Carrara.
- Fix the hair.
- Create ground plane, sky, lights. Animate the camera a bit at the end.
- Render to TIFF
- Assemble and add music in Premiere Elements 4.

Problems noted:
- The hair element didn't import properly into Carrara - it came in with no texturing as a solid grey mass. I had to remove the hair and reload a copy of the asset from directly within Carrara, reparenting to the Aiko3 figure. This problem suggests a bug in either the Daz Studio method of exporting parented hair assets, or Carrara's method of importing them.

- Daz Studio uses z-axis forward, the standard for most 3D programs, including every single Autodesk tool that I'm aware of. Carrara, however, uses z-axis up, which is much less common. The result is that any attempt to load my Daz-friendly BVH files directly onto a figure in Carrara results in what are essentially garbage keyframes on the hips, for both rotation and translation. This is a major failing on the part of Daz, which simply should not be shipping two separate 3D products that use an inconsistent xyz view of the world. Different xyz axis orientations is a classic situation that often makes it nearly impossible to move assets and scenes smoothly from one 3D application to another.

I could understand this difference in axis orientations if Carrara were a new acquisition for Daz, however Daz purchased Carrara several years ago and has now had at least 2 release iterations (versions 7 and 8) to standardize on the industry-standard z-axis-forward orientation.

However, in general the process of shifting a scene (with keyframes) from Daz Studio to Carrara was cleaner than most export-import processes I've seen, including Autodesk fbx, which has traditionally been a mess even when you're moving from one Autodesk product to another.

The cgspeed Daz-friendly release of over 2500 BVH files is now available

2010-07-30T23:18:00.001-07:00

I'm pleased to announce the most recent BVH conversion of the Carnegie-Mellon (CMU) motion capture library. The new conversion is for the Daz (daz3d.com) 3rd-generation and 4th-generation characters such as Aiko3, Aiko4, Victoria3, Victoria4, Michael3, Michael4, and so on. Over 2500 separate BVH files are available, with an index. You can get the files right here at cgspeed.

Other people have previously done releases of the CMU motion dataset for the Daz characters or "for Poser", however sometimes there are significant problems using these earlier releases. This new "official cgspeed.com" release was created via a full retargeting in Motionbuilder, and is designed to import seamlessly into programs like Daz Studio when you use any of the Daz gen3 or gen4 characters.

Two training videos are available (in HD):
Part 1: www.vimeo.com/13777740
Part 2: www.vimeo.com/13777422

This Daz-friendly release is actually two separate releases: a "primary" and a "secondary" release. The difference between these two releases is explained in the second training video, and in the full READMEFIRST file of the primary release. Generally you should start with the primary release, and only get the secondary release if you think you need it -- see the second training video for an example.

Thanks to John Lukasiewicz for doing the initial heavy lifting on the Motionbuilder Python script that enabled this release. No thanks to Autodesk for having bugs in their BVH export code, which I had to correct around. :-O

AFFILIATION AND DISCLAIMER: I (Bruce) am not affiliated with and don't speak for Carnegie-Mellon University, Daz Productions, Autodesk, or Smith Micro, and they don't speak for me. :-)

Animazoo Gypsy 7 announced - now under $10K

2010-07-24T12:10:00.001-07:00

Picture: the Gypsy 7 suit (on the right). The left suit is Animazoo's
inertial-based system, so ignore it - that's not a Gypsy 7

I received an announcement today about the upcoming Gypsy 7 motion capture suit from Animazoo. This is the successor to the Gypsy 6. Some official information is at:

http://www.animazoo.com/index.php/gypsy-7
http://www.animazoo.com/media/pdf/Animazoo_GYPSY7_Brochure.pdf

Here's a 3rd-party page on the predecessor Gypsy 6 system, for comparison:

http://www.est-kl.com/products/motion-tracking/animazoo/gypsy-6-technical-specifications.html

Gypsy is a potentiometer + inertial based system rather than a marker-based system. So the big advantage is you don't need cameras. Another advantage is that you're not theoretically limited by a capture area. The big disadvantage is that the performer's motion freedom is limited by the suit - you're not going to be doing complex martial arts demonstrations in one of these.

The huge difference from the previous model is the 3x (!) drop in price. A Gypsy 6 full-body system was about $25K, but Animazoo is claiming an entry price of $8K for the Gypsy 7, before software drivers. So a Gypsy is now in the same under-$10K price space as a Naturalpoint camera-based system. The Motionbuilder driver set is an additional $1000.

An email exchange with the U.S. distributor indicates that Gypsy 7 is a USB-tethered system. The Gypsy 6 could be run in wireless mode. The tether seems to me like a significant disadvantage - you don't want your performer to have some long USB cable snaking off the back of the suit to your computer for the data capture. One great thing about wireless no-camera mocap systems has always been that you can dump a performer into the middle of a field, or a parking lot, or whatever, and get data -- your capture area is the largest public space that you can find. So it's unfortunate to lose that for the Gypsy 7. I wonder if it would be possible to trick up something via consumer-grade wireless USB hardware.

From the pictures online, I wonder how well the suit is going to pick up full head and neck rotations. That looks like a solid bar pointing up the spine towards the head. But until somebody gets to experiment with one of these, it's hard to tell.

Disclaimer: I don't work for Animazoo, nor do I resell their products.

CMU motion capture BVH release for Daz characters is in progress

2010-07-07T21:27:00.001-07:00

Here's a 60-second test video that I made with a BVH file from the upcoming Daz-friendly conversion of the Carnegie-Mellon BVH dataset. I'm working on a new conversion which will complement the existing motionbuilder-friendly and Max-friendly releases. There has been at least one previous Poser-oriented conversion of the CMU dataset in the past, however investigation suggests that it has quite a lot of problems with bad joint rotations. This new conversion should avoid the problems of previous releases.

Blender for fan anime

2010-06-23T19:16:00.000-07:00

Blender (with After Effects) comes of age for toon-shaded, anime-based fan animation. Related sites include a brief BlenderNation article, and the creator's own site. Note the use of physics-engine-driven hair (both videos) and cloth (second video). Note how the hair and cloth are treated as fully-qualified objects which collide with the body geometry instead of intersecting it.

HapiLoli from tomo on Vimeo.

ルイズ／恋愛サーキュレーション from tomo on Vimeo.

Dancing 3D avatar software

2010-01-18T11:30:00.000-08:00

While it's not going to be iClone any time soon, the free software Mikumiku Dance appears to allow a fair amount of animation capability if your thing is making dancing 3D anime-style avatars. The software is originally Japanese, however an English release is available and several training videos, available at the same link, have also been subtitled in English. The software incorporates the Bullet physics library, presumably for the cloth and hair simulation on some of the models.

Other potentially useful sites:
Miku miku dance wiki
Miku miku beat

When you combine the software with live camera tracking and compositing software, you can end up with results like this video:

What the hobbyist animation community needs

2010-01-18T11:02:00.000-08:00

I recently received email from somebody in the animation software industry who wondered if I had any comments about the capabilities that hobbyist animators needs. Below is an edited version of my response.

1. Animation software

Hobbyists need an animation application with the capabilities of 3dsMax at a pricepoint of $500/year or less, not the current $4000 startup + $500/year maintenance approach, with workstation lockdown via licensing keys, that Autodesk presently charges. Fundamentally Autodesk is a problem for hobbyists because Autodesk has zero interest in the hobbyist market.

The three lead candidates for an animation app that might meet people's needs sooner rather than later are Daz Studio, which seems to be adding some rudimentary animation tools, but has a long way to go; iclone 4; and Blender 2.5. However iClone's business model is focused on selling their own proprietary content packs, not on importing Daz or other characters or clothes. Blender has not been a solution to date for a variety of reasons, including a user-hostile interface that doesn't match anything else in the industry; lack of a full scripting API which makes it impossible to write simplification tools; and a lack of usable .fbx import. Blender 2.5 claims to fix some of the UI problems, however version 2.5 is only in alpha and won't be in a less-buggy release for another 6 or 9 months or whenever they feel like getting around to it.

2. Motion capture

Hobbyists need a prosumer motion capture system that produces clean, professionally-usable capture data, can be built using high-end consumer-grade webcams, run at home on a laptop, for a total system cost of under $2000. A good NaturalPoint system presently goes for $10K, the cameras are proprietary, and they have a lifespan of perhaps 2-3 years until the next generation of cameras comes out. NaturalPoint is obviously much better than Vicon, which is completely out of hobbyist range, but fundamentally you're not going to see hobbyists doing home mocap-for-animation until the price point drops by another factor of 5 to 10, down to under $2000 including the cameras. There's a chicken-egg problem here: NaturalPoint has to charge $10K for their software + cameras because the market is small. If the market were 10x larger because it expands to include hobbyists, they, or somebody with a similar system, could drop the costs.

I don't presently have a lot of faith that low-end markerless systems such as IPI's Desktop Motion Capture, which maxes out at 4 cameras, will produce capture datasets that are realistically usable without huge amounts of cleanup.

3. Very easy content portability

Hobbyists need very easy, 100% reliable content migration and import from content sold for hobbyist applications like Poser and Daz3D into animation engines. All of the Daz characters should be trivially importable into Blender and all Autodesk applications (and C4D and LightWave), with rigging, skinning, and morphs intact. I'm not interested in static .obj file export/imports where I'm told "now just go skin it and rig the mesh yourself in your animation software" -- I've spent hours trying to skin Daz characters within 3dsMax and the results are still unusable, because skinning is a specialty skill. I don't want to have to build up my skinning skills to the level of expertise required to skin a 120,000-vertex character for semi-professional use -- I've already purchased the Daz content which is skinned within Daz3D, it should work in my other 3D applications as well. Nor I should I need to spend 10 hours reverse-engineering what's going on with the textures to make the eyeballs look right, which is what you have to do with Maya or Max today when converting a Daz figure.

I should then be able to purchase content at Renderosity or Daz -- at their consumer-level prices of roughly $5-$30 for some clothing items, a vehicle, or a set -- and trivially load it into my commercial 3D application, again with morphs and textures intact. I should be able to buy clothing sets for the Daz or Poser characters and easily pull them in to the animation software to use on the figures. This process works in Poser just fine, but Poser isn't an animation tool, it's a still-figure posing tool that isn't usable for serious animation work.

In my experience, none of the tools or formats that claim to support 3D asset migration work properly or cleanly. Not once have I seen a Collada export-then-import process successfully maintain skinning, rigging, textures, and morphs. FBX import/export tools don't work either -- even Autodesk can't get this right within their own product lineup; they release a new flavor of FBX exporter/importer every 12-18 months and you have to synchronize your FBX tool flavors to have a chance of even making a Maya-to-Max or Max-to-Maya conversion work. Pcharacter2FBX, brokered through Daz, comes close to performing as advertised, however even it does things like mess with the underlying bone structure of Aiko3, and it doesn't work on the 4th-generation Daz characters since it hasn't been updated in years.

Max also is severely limited because it only supports 100 morphs per mesh geometry. Even Aiko3's full morph set is around 200 morphs, and Daz V4 is probably many more than 200. So right now it's not even possible to pull the Daz 3rd-generation characters into 3dsmax and preserve all their morphs without an expensive 3rd-party morph-extender plugin.

I've also found that when you import .obj files into Maya, none of the file-based textures come in, so you have to replace them by hand. That's another example of the import-hostility of current software. In the Autodesk world if you're not using 100% Autodesk and .fbx, you're expected to have an on-staff Python/PERL/MEL programmer who can write fixup scripts and design your data-migration pipeline for you. No hobbyist animator has the time to do that, nor should they have to learn Python.

Very low-cost portable sound booth

2009-05-07T18:57:00.000-07:00

Good audio is critical for any video work - in fact, many creative professionals have noted that more video projects end up unsuccessful due to bad audio than due to any other reason. Traditionally the solution to getting good voiceover audio has been to rent studio time and record in a soundproof booth, however some voiceover artists have been getting creative by making low-cost, portable, mini soundbooths that shield just the microphone. Here are a few links to what's essentially the same concept: a collapsible cube with soundproof foam on 5 of the 6 sides.

- Build a $21 Portable Vocal Booth from O'Reilly Digital Media blog
- DIY Portable Recording Studio from Jake Ludington
- Portabooth info page from Harlan Hogan

CMU 2500-motion database released in 3dsMax-friendly version

2009-05-06T17:34:00.001-07:00

I'm pleased to announce a 3dsMax-Biped-friendly BVH release of the Carnegie-Mellon University (CMU) motion database. Last year I released a Motionbuilder-friendly version of the data, but because Motionbuilder tends to be a specialized tool and I recently switched to Max anyway, I decided that instead of having a life, I'd invest some research and programming time to convert the dataset to something easily usable with Biped. I've also recorded a training video to show how to import the files and retarget them to another Max Biped.

In the wake of the initial BVH release of the Motionbuilder-friendly version in 2008, various people have done different types of conversions and published them. So there's now a Poser-capable version, and apparently also versions saved as Collada (.dae), Maya Binary (.mb), iClone (.vns) and more, though I haven't tested any of these. I have links to the various releases on the top-level cgspeed.com motion capture page.

The motion data is free and available to use for any project. There is no use license, i.e. the data isn't under Creative Commons or a similar limited-use license.

WHERE TO GET THE FILES:

1. The 3dsMax-friendly BVH release, main page, with download links:
http://tinyurl.com/dk8n5z
OR http://sites.google.com/a/cgspeed.com/cgspeed/motion-capture/3dsmax-friendly-release-of-cmu-motion-database

2. The top-level cgspeed.com motion capture page:
http://tinyurl.com/d6vead
OR http://sites.google.com/a/cgspeed.com/cgspeed/motion-capture

3. Tutorial video:
http://www.vimeo.com/4449838
OR download the full-size version from the 3dsMax BVH links page mentioned above.

Advantages of this release over the original Carnegie-Mellon data:

- Very easy and fast to use with 3dsMax Biped.

- Joint renaming: As many joints as possible have been renamed to be compatible with 3dsMax's BVH import capability for Biped.

- Easy retargeting: since 3dsMax Biped allows biped-to-biped retargeting even across different Biped skeleton setups, it's now possible to use the CMU motion data with arbitrary Biped skeletons, at least within 3dsMax. This is a huge win, because it means that you don't have to create a skeleton whose bone setup matches that of the original CMU data.

- Base pose modification: 3dsMax requires an unusual skeleton BVH "base pose" aka "zero pose". For the 3dsMax-friendly release I've adjusted the underlying dataset to put the zero pose into the appropriate arms-down position. Numerous adjustments to the keyframes on the arms were then required to handle the change in rotation axes that happens when you shift the underlying BVH zero pose.

- Hand-to-wrist keyframe migration: This 3dsMax-friendly BVH release shifts keyframes from the "hand" joints onto the "wrist" joints, making this data available to Max. 3dsMax BVH import doesn't support both "wrist" and "hand" joints, so without this datashift, the wrist rotations of the original capture would end up lost.

- Index files: The release includes consolidated indices that list the motion filenames and their descriptions. Both spreadsheet and word processor friendly index files are available.

Motion capture report from GDC09, San Francisco

2009-03-28T11:24:00.000-07:00

I attended the Game Developer's Conference (GDC'09) in San Francisco on Friday March 27 so that you didn't have to. So if you missed GDC this year, this is your lucky day. I don't develop games, and have no plans to develop games, however GDC does have some motion-capture vendors on the exhibit ("expo") floor, plus a few other exhibitors who sell software/hardware toys that are potentially useful for hobbyist animation.

Since my interest is strictly in mocap systems and in products that make hobbyist animation faster, this is a highly-focused writeup that doesn't touch on the vast majority of vendor content from the expo floor. Many of the exhibitors were game middleware vendors, testing/QA service companies, national agencies trying to convince game developers to relocate to Canada / Germany / Scandanavia / Singapore, or schools that offer programs in computer animation. However there were also three full-body mocap vendors who attended with full demos, so I paid my $75 student entry fee (I'm a card-carrying very-part-time student) and hit the expo floor.

Here I have a heavy focus on NaturalPoint, because they're realistically the only mocap company with a product in the hobbyist range -- and even that's pushing it, depending on what you mean about disposable income levels when you say "hobbyist". I really should have asked the reps at Xsens and PhaseSpace what their system prices are, however I didn't, and neither of those companies is forthcoming in print about its pricepoints like NaturalPoint is. I'm fairly sure that when I've researched their prices via web searches, they've been significantly higher than the $5K-$10K pricepoint of a NaturalPoint system.

Interestingly, the 3 mocap vendors at the expo all use different technologies (optical passive marker, LED active marker, and inertial). All of them had live demos running with an actor or actress wearing the appropriate suit, and the data streaming to one or more live video monitors with a reconstructed virtual actor on-screen.

NATURALPOINT:

There's actually quite a bit of incremental news here, though apparently they haven't published it much on their web site yet.

The biggest news is that the new cameras are coming. This is all in the printed glossy materials they were handing out, so it's presumably all public at this point. NaturalPoint is about to roll out the "V100:R2" camera, which is the second generation camera that will replace their current "FLEX:V100" series. The R2 looks very similar to the R1, however it's got twice the FPGA programming space internally, so NaturalPoint will have more flexibility with what they can do with the camera's internal software over time.

The R2 camera will also support interchangeable lenses, which turns out to be a huge win. I don't believe that the original first-gen camera allows you to change the lens, though NaturalPoint will presumably clarify that as they get more documentation online. The first-gen camera has a 45-degree field-of-view on the camera, but the R2 camera will support a 60-degree field-of-view lens if you choose to install that lens type. The 60-degree lenses allow you to increase your capture area without increasing the size of the room where you're capturing. There is presumably some cost in terms of accuracy, since the cameras are still only 640x480 devices, however the brochure makes the win really clear:

-- With the original cameras, to capture a 10'x10' square space, you pretty much needed a 20'x20' room. (At least, this is what the diagram in the literature suggests). The NaturalPoint material estimates that about 20% of the volume of the 20'x20' room is usable as actual capture space in this original configuration.

-- With the R2 cameras and 60-degree lenses, to capture a 10'x10' square space you only need a 16'x16' room. NaturalPoint estimates that in this mode, about 34% of the volume of your room is usable as capture space. So that's about a 50% increase in your capture volume (from 20% of the room to 34% of the room) just by swapping out the lenses.

So we can mark NaturalPoint down as probably one of the only vendors which is improving its engineering to DECREASE the amount of capture workspace that you need. This clearly is reflective of their customer base, which presumably has a heavy component in the "garage or living room hobbyist" demographic like me. Since I don't have a 20'x20' open room in my home (not even my garage will do that for me), the R2 camera now opens up the serious possibility for me of getting close to the full theoretical 10'x10' square capture space at home.

The new cameras will cost slightly more than the old -- it looks like it will be $600 rather than $550 -- however the base price for a 6-camera setup will apparently still stay at $5000, so the actual cost for a new 8-camera or 12-camera setup won't go up by much.

The second piece of news is the 3dsMax streaming plug-in, which NaturalPoint has mentioned in their online forums. They also have a sample video up on the web site that shows it in use. The plug-in works much like their MotionBuilder and Daz Studio plug-ins: you still need to run their capture software "Arena" on one PC, but now you can also fire up the Max plug-in and stream data live directly into 3dsMax, where you can target it directly onto a Biped object. Since I very recently decided to dump Maya for Max, this is also positive news for me.

The Max plug-in was running at the GDC'09 booth, so I can confirm that it was receiving data from the live mocap actor and sending it straight onto a Max biped.

It's not clear if this is officially announced or what the timetable is, however NaturalPoint suggested that they're working out a way to trick up the camera "sync" signal onto the USB cabling, which would mean that you no longer need to run two sets of cables to each camera, instead you'd just run the USB cable. So that will be a win for cable control and setup time once that feature is available. Right now the glossy literature still shows sync cables, so those must still be a necessity. I noticed from their web site that they've also got a new "second-generation calibration wand" which is already part of the calibration kits - this new wand is extensible up to about 5' long, so you don't have to wave a little 18" pixie stick thing frantically in front of you to calibrate the space anymore.

Now let's do a dive into some of the geek details that you might care about if you're thinking of buying one of these systems.

-- The GDC09 NaturalPoint demo booth used a 16-camera truss-mounted setup in a total area of about 15'x20', with a performance stage and capture area of about 9'x9' (probably 10'x10') in the center. This setup allowed easy single-actor capture and allowed them to attempt dual-actor capture in the afternoon -- more on that below.

-- One thing I noticed about their setup is that one of the lower corner cameras was NOT mounted directly below the upper, rather it was mounted an additional 18" or so backwards, which presumably helped the capture area a tiny bit. I confirmed with the salesrep that it is NOT necessary to have your lower cameras mounted directly below your upper, even though the online diagrams suggest that this is how you should do it. You have some flexibility in the horizontal positioning of the cameras. This is helpful information for somebody like me with weird pipes in my garage which will occlude upper cameras but not lower, for some corners of the garage. It sounds like I can mount each individual camera as far back in the corner of the garage as possible, as long as there's no occlusion.

-- Arena (the NaturalPoint software) was running on one dual-core laptop and I'm 98% sure that the Max plugin was running on a separate laptop.

-- In the afternoon I stopped by again and they were attempting to demo a 2-actor capture, however the system was having problems tracking one of the actors -- limbs were often going out of joint or the figure would turn sideways or such. The explanation given is that for 2-actor capture you really need a quad-core CPU, and that they were overloading their dual-core with all the cameras and the markers to solve. This was the behavior I observed as well -- it looked like the displayed point clouds were sluggish to update on the large Arena software display screen. So the rule of thumb seems to be: use a PC with at least 2 CPU cores per actor. If you ever wondered why in the world anybody would need a quad-core laptop, the answer isn't "so that you can run Microsoft Word REALLY REALLY FAST", it's "because I want to do 16-camera two-actor motion capture." You knew Dell and IBM were selling those expensive laptops for a reason, right?

- I asked sales if it was possible to run both the Max plug-in and Arena on the same PC, and got a nervous response that suggested that CPU matters a lot in this situation. The recommended config was Arena on one PC, Max on another, then you stream the data across your local network from Arena to Max. This is probably how I'd set it up anyway, so that's not a problem in my case.

- They also were slightly nervous when I told them that my living room is only 13'x14' -- the response was that yes, I'd get a capture volume, but it wouldn't be anything close to 10'x10'. They didn't mention the upcoming rev2 cameras with the new lenses at the time -- that was a later conversation. It still sounds like my 20'x15' garage is going to be a better choice -- in fact I had something of an extended "garage versus living room" discussion with them, the outcome of which was that I vowed to re-measure my garage size when I got home.

- High ceilings matter. The preferred ceiling height for camera mounting seems to be about 10', with current OptiTrack online videos suggesting that your high cameras need to be AT LEAST 8' high. Unfortunately, many houses including mine have 8' ceilings. It's probably not a showstopper, however it will limit my ability to have an actor stand on a footstool or jump in the air or such.

- The full Arena system is locked down only to the dongle, and is NOT locked to any particular CPU or machine. This means that unlike Autodesk products or similar node-locked software, you can choose to sell your entire Arena system to somebody else if you decide that mocap isn't for you anymore. NaturalPoint confirmed that this was fine, they have no issues with people reselling their systems used. To me this seems like a fairly significant selling point -- if you put down $8K for a NaturalPoint system and then you decide in a year that you really don't want it, you can probably recover a good portion of your original cost if you find a good buyer.

- The Max plugin (I believe it's still in beta) apparently doesn't presently support anything other than a single Biped in the Max scene, i.e. it sounds like you can't yet add any geometry to the scene so that your mocap actor can touch virtual walls, lean against a virtual table, etc. They suggested that they hope to eliminate that limitation in future revs of the plug-in.

PHASESPACE:

PhaseSpace uses an active-LED marker technology wired into the usual mocap suit, so when you're tricked up in one of these, you look like a glowing red Christmas tree. They list a camera resolution of 3600x3600 and a capture rate of 480 Hz, which is one reason why I'm fairly certain that this is an expensive system. (By comparison: NaturalPoint uses 640x480 cameras at 100 Hz.) PhaseSpace also sends the cameras into a proprietary capture box, which is actually a dual-core PC running their software. So every PhaseSpace mocap system ships with its own custom PC.

The win with the active markers is that your software should never accidentally swap markers around, which means less cleanup of your data after the capture.

PhaseSpace is the one vendor who offers an optional glove for the mocap suit that can capture finger position. The other vendors at GDC'09 don't have a way to track individual fingers -- they can only track the position of the full hand, and then you have to animate the fingers on your own. If you add the glove to the PhaseSpace system, you get extra active markers on the fingers so that the system should be able to know how the fingers are bending.

The recommended usage pipeline is to send the data into MotionBuilder -- in fact it sounds as if the PhaseSpace system is simply collecting a set of (x,y,z) datapoints for you on the proprietary PC that ships with the system, and then you stream those datapoints into MotionBuilder and use MotionBuilder to do the bone solve. This approach appears to be somewhat less than what NaturalPoint does with its software, since with NaturalPoint you can stream directly into Max or Daz Studio, and I guarantee that Daz Studio doesn't have a built-in point cloud solver that determines how to position skeleton bones using a set of tagged (x,y,z) coordinates.

I asked if it would be possible to use just the PhaseSpace glove without the suit, and the answer was one of those "in theory, yes" answers -- in theory yes, if I write my own intermediary software to take their datastream and send it into my preferred 3D application onto a hand rig. However they don't sell a plug-and-play hand-capture-only solution -- the glove is really meant to be used with the full suit.

Price speculation: inition.co.uk lists an 8-camera PhaseSpace system with control/capture computer and accessories at 33,590 British pounds, so it looks like we're talking easily in the US$50,000 range for a PhaseSpace system. Inition does list a lower-end 4-camera system for a mere 21,000 pounds.

XSENS:

Xsens sells the "MVN" (formerly "Moven") inertial motion capture system, which is derived from their earlier work on biomechanics and medical motion capture. The technology here is very different from the other two systems, and has some advantages and disadvantages. An "inertial sensor" is an accelerometer, the same type of sensor built into the Wii controllers. It can detect changes in acceleration and you can then integrate these to get velocity and positional CHANGE, however there's nothing in the sensor that can give you absolute (xyz) position in space.

Their demo setup featured a one-actress system streaming live into MotionBuilder.

Some differences between the Xsens/MVN system and the other two systems shown at GDC09:

-- No markers. The system uses strictly inertial sensors, so you don't have to stick markers onto the suit. The suit does, however, have small-but-obvious inertial sensors embedded in it -- these appear as rectangular lumps on the suit.

-- No cameras. Since you have no markers, you have no cameras to capture the markers.

-- Huge capture space. Since you don't have to stay within a "capture area" bounded by cameras, your actor can go wandering all over the place and the system will still capture it. You're only limited by the range of the WIFI transmitter built into the suit, and that can go up to 100' or so. The sales rep showed this by having the actress wander around outside of the booth space, and the system captured the motion just fine.

-- Position drift. Since the system uses only inertial sensors, it has no way to confirm the absolute xyz position of the hips. The sales rep mentioned that although the mocap actress could walk all the way down the expo floor and back, and the system would capture all of that motion, there would probably be a drift of at least a few inches in position by the time she got back. Whether this makes a difference for an animation pipeline probably depends on the pipeline and the need.

The Xsens system is biped-only, and streams the data into a custom application takes the accelerometer data, applies specific knowledge of human biomechanics, and uses it to produce the human motion on the screen. In other words, there's a heavy amount of "computer algorithmic assist" going on -- the software isn't actually tracking the position of every joint, rather it gets information about the motions of SOME joints, and it uses it knowledge of how the human skeleton is wired together, and how human beings move, to figure out how the skeleton should move.

The fact that the system doesn't know the xyz position of the hips leads to some unexpected behavior when your actor climbs stairs or a stepladder. Since the system can't track the hips height, it has to make an assumption, which is: "the ground is always at the same level, and the hips will always settle to the same level after a stride." It means that if your character climbs a ladder, the Xsens system will read each step as if your character simply took a large step up, and then settled back down to the floor. You have to manually go into the software after the capture session and tell it "no, the floor moved to a higher level at this point" on the relevant footstep keyframes as your character climbs up or down. I'm a bit surprised that Xsens hasn't added a hips-height capture mechanism to their suit to solve this problem, however that would then require constraining the mocap actor to an enclosed space near to the height sensor, much as an optical system requires you to stay near the cameras. (Didn't Animazoo solve this problem with an inertial gyroscope or something, in the Gypsy system?)

Price speculation: a thread on cgsociety.org from early 2008 estimates a pricepoint of $50,000 for the system.

-------------------

SUMMARY: NaturalPoint Arena remains the only full-body motion capture solution that I'm aware of that falls into the under-$10K category. The other vendors at GDC'09 are targeting small and mid-sized game studios who want to cut development costs by doing some mocap in-house rather than outsourcing all of their mocap needs to a specialty studio like Red Eye, but who don't want to pay $100K or more for a Vicon system. They can presumably be good choices for a 20-person studio that brings in a few million dollars a year in revenue and has a budget for R&D, but $30K to $50K for a mocap system just doesn't cut it as a pricepoint for the hobbyist. With the upcoming introduction of the new 60-degree camera lenses, NaturalPoint is continuing to make enhancements presumably aimed at helping the "living room animator" do a better job.

WEB SITES:

Naturalpoint: www.naturalpoint.com/optitrack
PhaseSpace: phasespace.com
Xsens MVA: xsens.com/en/general/mvn

------------

Disclaimer: I don't work for any of the companies mentioned here, and none of them asked me to write this show report, no money is changing hands, etc.

3DCG for the hobbyist - a short reference document

2009-01-10T21:43:00.000-08:00

I've made some edits to a previously-published "one-pager" document which provides a large number of links to software and tools for 3DCG work. The document is titled 3DCG for the hobbyist - useful products and links. It was originally designed to fit onto both sides of a single sheet of paper as a handout, but now you can read it here online.

News items

2009-01-10T17:35:00.000-08:00

Here are a few news items relevant to the hobbyist CG community:

- Craft Animations has released free lite versions of several of their Maya and 3dsMax camera tools, including Craft "4-Wheeler Free", "ObserverCam Free" and "Airplane Free". They have a press release online with more information. It would be somewhat cute to speculate that the Realtime Camera Motion Recording in Maya tutorial -- which shows how to do ObserverCam-like camera recording for only the price of a Spacenavigator -- that I posted to this site back in June had some influence on Craft's decision, however I have my doubts.

- Also from Craft, but not free, is the recently-announced virtual camera product released jointly with Gamecaster: a physical camera mount and camera control which links directly into 3dsMax and Maya. The product combines Gamecaster's camera-control hardware product with Craft's software. The press release is one of those "contact us for pricing" sorts of releases, so the product probably isn't cheap. However, this is the first time I've seen such a hardware/software setup commercially available for animators -- previously I'd only seen similar tricked-up systems shown in "making of" mini-documentaries associated with movies.

- On Nov. 28 of last year, Antics Technologies announced the discontinuation of machinima software Antics3D, which was available in both a free and paid version. I was aware of Antics3D but had never explored it, since it didn't look as robust as iClone version 3.

Incidentally, if you're looking into purchasing iClone, one way to do it and save some money is via the Daz3D Platinum (membership) club -- iClone3 Pro is available in its download version for just under $140 at the moment to Daz Platinum Club members, and Daz once in a great while puts even that price on sale. (Disclaimer: this is not a paid announcement, nor do I work for Daz...)

Poser-friendly conversion of the CMU BVH files

2008-09-25T21:54:00.001-07:00

Mike Sutton of mojodallas.com has taken the BVH conversion release of the Carnegie-Mellon motion capture dataset and run it through his own set of scripts to make it more feasible to use the motions in Poser. You can find the main page that describes his conversion work and the download links at mojodallas.blogspot.com. The files are presently posted at sharecg.com.

Announcing BVHplay, a free BVH animation player

2008-09-19T00:05:00.000-07:00

I've just released BVHplay, a free, lightweight, Python-based, multi-OS, open-source playback utility for BVH animation files.

To learn more about it and/or download it, head on over to the main BVHplay page. BVHplay is also available from Sourceforge, however I find the Sourceforge file-selection interface to be somewhat confusing, so most people will probably prefer to download BVHplay from this site.

Followup to CMU BVH files release

2008-08-30T11:51:00.000-07:00

In the weeks since I released the BVH conversion of the Carnegie-Mellon motion files in July, I've received a variety of email inquiries from people who quite sensibly would like to get the files to work with their preferred animation software, which typically isn't MotionBuilder. The email I've received has been interesting because it suggests that people are using a variety of applications. So far I've had multiple email inquiries from Poser 7 and/or Daz Studio users, multiple inquiries from iClone users, one from a 3dsMax user, and one person way out there at the high end using Endorphin.

Unfortunately the answer at this moment in time is "there is no general-purpose, low-cost, easy method of converting BVH files from one character rig to another". Although today you can take the BVH-converted Carnegie-Mellon motion files and use them in a professional animation retargeting program like MotionBuilder, many people using consumer or prosumer-level animation software aren't having a lot of luck making coherent animations using these files.

"Why is this so?" you might ask. It's a good question. The answer is that any given BVH file is designed to apply to a specific skeleton that's been set up (rigged) in a specific way. If you use the BVH file with its original skeleton, the animation works. If you try to use the BVH file with some other skeleton, or with the same skeleton that uses a different joint rotation setup, you'll typically get crazy animation - the arms or legs fly all over the place, for example. This isn't the fault of the BVH file, it simply means that the BVH file was never designed to work with the skeleton or rig of the new character.

The general problem is called "motion retargeting" or "animation retargeting" and is an active area of research, particularly for complex or varied skeleton setups, in the professional animation community - see for example this 2008 SIGGRAPH paper which discusses animation retargeting for SPORE.

Specifically for SecondLife use, one program that you might try is BVHacker, however I haven't experimented enough with this to know how well you can apply it to the Carnegie-Mellon BVH conversion release that I've done.

As always, if you come up with a workflow that works, something that "makes hobbyist 3D animation easier, cheaper, faster", feel free to email it my way (see the "about the editor" sidebar here for my email address) and we'll see if we can turn it into a blog post here at cgspeed.com.

- Bruce

Announcement: cgspeed.com release of over 2500 free BVH files

2008-07-21T21:03:00.000-07:00

I'm happy to announce the free release of an enhanced BVH conversion of the entire set of 2548 human motions from the Carnegie-Mellon Graphics Lab Motion Capture Database. This is a large set of professionally-captured human motions of a wide variety of types, suitable for use in animation software, which has not previously been available in BVH format.

Please see the announcement and download page for the full details.

New tutorial: Installing Maya 8 on Fedora Core 8

2008-06-29T17:16:00.000-07:00

I finally wrote up my notes about how to install Maya on Linux (Fedora Core), with some bonus notes about how to install drivers for an Nvidia video card. You can read the tutorial here.

New tutorial: Realtime camera motion recording in Maya

2008-06-27T00:05:00.000-07:00

My first tutorial is up! Have a look at "Real-time Camera Motion Recording in Maya for under $50: Using a SpaceNavigator for Maya Camera Flythroughs". I also recorded a 15-minute video tutorial that supplements the written tutorial - the video tutorial appears below. Some of the text in the MEL-script-edit part of the video tutorial is very difficult to read, so I do recommend that you also print the written tutorial and follow along.

Realtime Camera Motion Recording in Maya

Welcome to the site

2008-06-27T00:01:00.001-07:00

June 27, 2008: this site is now launched, and you can read the welcome and manifesto.