A 3D Curve Sketching System For Tablets

dominique_cimafranca writes "The Dynamic Graphics Project of the University of Toronto has released a pretty nifty 3D curve sketching system. Apart from the large drawing area, the tablet software looks very intuitive to artists. From the site: 'The system coherently integrates existing techniques of sketch-based interaction with a number of novel and enhanced features. Novel contributions of the system include automatic view rotation to improve curve sketchability, an axis widget for sketch surface selection, and implicitly inferred changes between sketching techniques. We also improve on a number of existing ideas such as a virtual sketchbook, simplified 2D and 3D view navigation, multi-stroke NURBS curve creation, and a cohesive gesture vocabulary.'"

Comments

Serial tautoloprone laminant curveoids

[Anonymous Coward]

I recall from my multiliniear calculus course that the fundamental zeroid of the Draper function is orthagonal in [n-1/n] hyperspace to the semi-Euclidean plane of the minimal Pascal rectangle. So if you point at one point on the tablet, multiple points are mooted when the gesture constrains pretensioning on its hypothetical "theta" axis. In other words, poo.

Re:Serial tautoloprone laminant curveoids

[sexconker]

That's the [N+1/N] you fool.

Re:Serial tautoloprone laminant curveoids

[martin-boundary]

Don't say that! Mr T-square is gonna open a can of Whoop Ass on us, for pity's sake!

Re:Serial tautoloprone laminant curveoids

[Plaid Phantom]

No, he'll just pity you. Pity you so hard you'll never see the light of day again. And then he will pity you even more for it.

Mr. T's pity is second only to a Chuck Norris roundhouse.

Re:Serial tautoloprone laminant curveoids

[Hucko]

Diamonds, rare metals

POP!

[John Pfeiffer]

I think my head just exploded into candy...

As an illustrator and 3d modeler, I must say, that is simply the most awesome thing I have ever seen. I would go so far as to say that it is 'insanely great'. I also just happen to be buying a Wacom Cintiq 21UX in the immediate future. FORTUITOUS!

Re:POP!

[lysergic.acid]

is that the tablet that was used in the demo video? that tablet + ILoveSketch is a pretty slick package. i'm looking to buy a tablet of my own, but i'm probably going to have to settle for an Intuos 2 or one of the other non-LCD tablets.

i've been doing graphic design for a few years, but i never got around to buying a digitizing tablet. i've been thinking about it for a while now, but this application demo has pretty much sold me on getting one. i just need to figure out which model to purchase. i'm wondering if i should get a 12x19" Intuos 2 (~$700), or a Toshiba tablet PC. what are others' experience with tablet PCs? are they a good substitute for a Wacom Tablet?

Re:POP!

[Fluffy the attack ki]

Some tablet PCs might be ok, depending on what sort of tablet screen it has, but the one I purchased had terrible pressure response and accuracy. In practice it felt like drawing with an old disposable Bic.

I now use an Intuos3 6x11", which I picked up for about $300. Excellent pressure response, no perceptible position jitter, and just a much better feel in general. I sometimes miss being able to turn the computer to get a better angle, but there are quick rotation controls in many drawing apps.

Re:POP!

[Pearson]

I don't think you are going to get a tablet PC that can touch a Wacom just yet. Also, I would suggest trying a Wacom before you buy, if possible. While many people think bigger is better (insert joke here), after actually using a Wacom, they usually find a smaller size fits their needs better. I've found the 6"x11" to be fine, while the 12"x12" was too big. The one you're talking about, the 12"x19" is actually about 25"x17", and almost 1.5" thick. That takes up a lot of desk space, and is probably too big for your lap.

Re:POP!

[John Pfeiffer]

Yeah, as far as I can tell, it's a Cintiq 21UX that's being used. ($2000!!!)

12x19? I didn't know they made an Intuos2 that big. Certainly seems like overkill. I have a 9x12 Intuos3 Special Edition ($500) and it's pretty big.

As far as I know, even tablet PCs using Wacom 'Penabled' tech lack some of the finer points associated with their graphics tablets. They really need to make a 'digital sketchpad', a tablet PC with all the refinements you expect from a high-end graphics tablet. (It needs to be able to use Intuos3 pens like the Cintiq!)

Re:POP!

[flewp]

In regards to a digital sketchpad, I would love one. I'm considering a Cintiq just because I don't know of any Tablet PCs that have both a great display (in terms of being a display and a drawing surface), and gobs of power required for intense graphics apps. And if Apple did it, well, the marketing just writes itself - they could call it just that, the Sketchbook, and it'd fit in with the naming convention of Macbook, iBook, etc.

Re:POP!

[compro01]

Every tablet PC I've seen uses Wacom's tech ("penabled"), just a somewhat detuned version of it to make the cost more reasonable. I have an HP tablet laptop (tx2500) myself. works great for writing out notes with diagrams, though as the above guy said, there's very little in the way of pressure sensitivity, maybe 3 levels of it, though I haven't had any problems with accuracy once I configured that.

Re:POP!

[SoupIsGoodFood_42]

Make sure you try a non-LCD tablet (and an LCD tablet, for that matter) before you buy one. I had a non-LCD Wacom tablet, but I never used it because I couldn't get used to it. I recently got a Wacom Cintiq 12WX and don't regret it. It is on the pricier side, but not as much as the high-end Cintiq 21UX. Wacom also have some that are a bit larger than the Cintiq 12WX for around the same price, but they don't have the extra pressure or tilt sensitivity or the handy touch sliders and buttons on the side.

I haven't used a tablet PC, but I doubt they would be as good as a Wacom if you want it for graphic design or art.

Re:POP!

[SoupIsGoodFood_42]

Pressure sensitivity can be a pretty big deal -- 512 levels offers functionality that can't be achieved with just 3. Also, the higher-end Wacom tablets also have tilt sensitivity. Usually tilt is used for doing calligraphy in Photoshop and such, but I imagine it could also be useful in a 3D app like this.

Re:POP!

[entgod]

Actually, I think tablet pc:s usually have 256 pressure sensitivity levels. Not sure about your tx2500 but at least my lenovo x61t does. The apparent lack sensitivity layers is probably due to bad configuration or cheaply built pen.

Re:POP!

[compro01]

Or it could be just me. My fine motor control sucks.

Re:POP!

[waztub]

As a tx2500 owner myself, I can testify that the official documentation says it has 256 levels of sensitivity.

Definitely not 3, from personal experience, regardless of official papers.

Re:POP!

[xouumalperxe]

And in the same train of thought, I got myself a Bamboo Fun a month or two ago (what can I say? Hobbyist on a student budget), and I've since figured that another important consideration is that you really want to make sure the tablet aspect ratio matches your screen's. You have the option between absolute mode and mouse mode (mouse mode behaves like a mouse, absolute maps the tablet to your screen absolutely), and absolute mode makes a lot more sense when horizontal and vertical movement have the same scale.

Could be useful for video game artists

[Dutch Gun]

This seems like it could be very useful in bridging the gap between concept art and a fully rendered 3d model. I'll have to remember to point this out to a few of my artist colleagues at work and see what they think about it.

Of course, I'll probably have to warn them to turn off the sound first. Quick hint to the developers of this cool little toy: Artists get nervous when when programmers start talking about "single view symmetric epipolar method" and other very complicated terms. If you've ever worked with artists before, you know you're starting to get too technical when the eyes start glazing over. I then know to take a step back and try to re-phrase in non-tech.

All you programmers are now thinking "but... that's exactly what it's describing", and I'll just put my hand to my head and sigh. Different ways of thinking.

Don't even get me started about trying to get in the heads of game designers.

Re:Could be useful for video game artists

[SnoopJeDi]

The shift in the nature of the software to take advantage of this doesn't really seem worth it, when you could just make two or three sketches of a concept at different angles, put those as the background in your modeling suite, and rough it out to the concept, then refine.

Re:Could be useful for video game artists

[seanalltogether]

I just showed this to my 2 roomates who do most of their art on tablets pcs. Their first reaction was "where do i get this", their second was, "what the heck are they saying". So good call on the technospeak confusing the hell out of them.

Re:Could be useful for video game artists

[mikael]

I would be interested to see a comparison of the speed of this system against the other different methods for artists to generate content. Some artists would create clay models of the character, have that baked, and then use a laser scanner or digital 3D pen to acquire a mesh. Others would draw sketches of the character viewed from different directions, then import these into a 3D modeller, set up three views and slowly tweak the control points until a match was achieved in a fourth view window (a 3D Expo tutorial). Yet others would just build the character from parametric surfaces, then convert to polygons and tweak the final vertex points.

Re:Could be useful for video game artists

[Dutch Gun]

It would be an interesting experiment, but the big variable - and I think one impossible to judge with any scientific methodology - is how each method helped or hindered the artists' creative process. If this tool allowed them to quickly try out new concepts that would only otherwise be available to a more freehand 2d process... it would seem to have some promise. It's a little hard to tell from a short video whether this would be useful, because we saw a few very limited examples with what was obviously a very expert user.

Even if this system ultimately doesn't prove itself in the market as a commercial product, I think it's likely pioneered some pretty creative interfaces that may *evolve* into something that does end up revolutionizing 3D computer modeling. Not to oversell it or anything, but it really does seem to have potential in the way it uses natural gestures and tries to intuit what the *intent* of the user is, rather than doing what most computer programs do - is accept *literal* input.

Re:Could be useful for video game artists

[catxk]

Quick hint to the developers of this cool little toy: Artists get nervous when when programmers start talking about "single view symmetric epipolar method" and other very complicated terms.

Just to make sure, to really make sure artists "get it", the tech stuff should be boiled down to something no more technical than "Hi, I'm a Sketch board"

Re:Could be useful for video game artists

[flewp]

Well, in terms of speed, this system could help cut down on the time needed to create the curves, but nothing else really. Simply drawing the NURBS curves isn't enough to actually generate a model, you have to use those curves to derive surfaces, which is rarely just a simple "one click" type of operation.

Define "released"

[Communomancer]

The Dynamic Graphics Project of the University of Toronto has released a pretty nifty 3D curve sketching system

I see a video and some links to bios and sample sketches, but no "released" software anywhere.

Re:Define "released"

[compro01]

I agree and can't find any available software anywhere referencing this project.

Re:So this isn't actually useful

[fuzzyfuzzyfungus]

Good point. After all, all 3D modeling software in the world is made by the same cabal. All time they spend on one project is stolen from another. In fact, every time somebody puts together a 3D modeling demo, another bug gets added to Maya, just to annoy you.

Seriously. You think Maya has annoying bugs, that's nice, go submit a support request or something. This is an utterly separate issue, similar only in that both of them are software for creating and manipulating 3D models.

strange

[deathguppie]

I thought it was going to be a 3d system for creating 2d drawings.. that would have been useful.

As it is (from a 3d artists point of view) this is just a more intense way of doing the same things that are already done with traditional 3d, and in fact comes nowhere close to what you can do with a sculpting program like z-brush.

If it gave me a 2d page I could turn and draw on like a real piece of paper.. that would be cool.. super cool.

... but it doesn't

Re:strange

[JamesTRexx]

I thought it was going to be a 3d system for creating 2d drawings..

And I was thinking of how it would improve my drawing of 3D curves... *facepalm*

Remind me to lay off the reading on ./ first thing after waking up early in the morning...

Cool.

[fuzzyfuzzyfungus]

The UI for open, save, delete, etc. seems gallingly stupid, just use the damn keyboard(yeah, flipping from corner to corner to turn pages will be realy intuitive when there are 500 of them...). The UI for drawing, though, looks amazing. Substantial amounts of the correct automatic stuff happening automatically, just really impressive translation of standard flat pseudo-3D sketching into 3D models. Most impressive.

Cool

[crashfourit]

This is so cool. I hope they opensource it.

Demo Presentation

[lamarguy91]

Looks like they are part of the presentation & demo sessions at the UIST (User Interface Software and Technology) being hosted by ACM next week.

More details here:

http://www.acm.org/uist/uist2008/

And a schedule of events:

http://www.acm.org/uist/uist2008/program/index.html

I hope to see additional project details and possibly some additional demonstration videos come from this event.

So...

[Ihmhi]

...what are the odds of getting a tablet laptop without Vista these days?

This seems pretty sweet, but I'm more interested in taking a tablet out with me rather than sitting behind a desk. I recall tablets hog more RAM than a usual OS, especially with vector graphics, so I shudder to think of trying to run this on a tablet under Vista.

Re:So...

[Nightspirit]

I'm running a tablet PC with 2gb of ram under vista and it works just fine. I've had the opportunity to run both XP tablet PC edition and vista on this same laptop, and didn't notice a speed difference between the two in day-to-day tasks. Everything pretty much opens instantly (with the exception of anything by adobe) and I never have any slow-downs or crashes.

I'd install linux on another partition if anyone can recommend a distribution supporting tablet functionality on a fujitsu.

Re:So...

[Ihmhi]

I'd be interested in hearing about that too, actually. I probably wouldn't be using it for games anyway and that's the only reason I use XP. (Yes, I know, you can run WINE etc. etc. but it's a moot point.)

Something similar...

[GrievousMistake]

That looks very nice!
I was looking at some similar stuff recently. There's an older app with some of the same gestures, called Teddy, (video here), which was further developed to Smoothteddy.

Here's hoping these interfaces will be further developed and reach mainstream, and that they will help artists that are good at drawing but bad at extruding, uv-mapping, etc. create some cool stuff.

pretty bitchin'

[rastoboy29]

if you can draw, it's pretty wicked looking.

Personally, I'll still with Blender, for now.

Which rocks, by the way.

Re:It has been done before

[CarpetShark]

If you had released what you did, and it had potential, someone might have finished it for you a year ago.

Rocks, as in "there's a rock in your shoe"?

[argent]

Blender rocks?

Blender's user interface is so bad that I can't imagine what the designer was thinking. Seriously. It's almost as if he hated end-users and decided that the only way he could express his hate was to make a program that appears superficially usable but caused as much pain and frustration possible when people tried to learn it.

Re:Rocks, as in "there's a rock in your shoe"?

[compro01]

That program doesn't have a learning curve. It has a learning cliff. But if you can manage to scale it, it's a very powerful and efficient program. The interface is both really bad and really good.

Re:Rocks, as in "there's a rock in your shoe"?

[ultranova]

That program doesn't have a learning curve. It has a learning cliff.

Learning Himalayas, with ferocious eagles attacking you every step of the way.

But if you can manage to scale it, it's a very powerful and efficient program. The interface is both really bad and really good.

Assembler is efficient, but that doesn't make it a good programming language.

Blender sucks. There's just no other way to say it. But then again, so does every 3D program. I suspect it has something to do with 3D artists wanting job security or something...

Re:Rocks, as in "there's a rock in your shoe"?

[Anonymous Coward]

I'm really not even a 3D developer, I just do reaally small projects just for fun and test some new features now and then but I really don't know what you're talking about. Blender has a great UI, it didn't take me more than a couple of days to get the hang of it and I think it's really productive.

It's also very much customizable and there is ongoing development to make it even more so in the next versions as the UI is getting a major overhaul.

Re:Rocks, as in "there's a rock in your shoe"?

[Skinkie]

But then again, so does every 3D program. I suspect it has something to do with 3D artists wanting job security or something...

Maybe you were joking, but if I think about this comment for more than one second you might actually have a point. But, the amount of architect/civil engineering students switching from *CAD to SketchUp cannot be ignored. Not because of the price but actually because of the ease of usage. So not every 3D program sucks, but maybe programs try to do all in one. Like Blender with all its modi, now you can produce a complete movie in one program... I saw the demonstration of this I love Sketch, and really, if this can bring SketchUp for Free Hand design, it actually has a good use to replace/enhance something. Things that were done in Illustrator, Xara, (Inkscape) in vector graphics can now be done in 3D, that sounds really cool to me.

Re:Rocks, as in "there's a rock in your shoe"?

[delt0r]

We get this a lot on /.

Intuitive for *anything* is really just what you learnt *first* when it comes to computers and 3D apps. I have used dozens of CAD apps and quite a few graphics apps (like blender) and I really don't think the criticizim that folk from the other "way" give is all that valid. Its like the DOS/windows wars, or worse Mac/PC crap. Each method/system/gui/workflow has strong points and weak points.

These things are Tools and gui are so far away from "intuitive" that UI experiment design should not be bogged down to some default standard. 3D is hard for most people and I have not found any one program all that much better or worse than another (but then I did do old fashion draughting for a while).

I find blender to be the best for mesh editing out there for the work I do. But at the same time IIRC there is work on customizable UI that will allow for a "mayer like" UI plugin. I think this is the future of UI for many apps: customization. That way it does not need to be a one size fits all. It can be the fast UI with non overlapping windows that I really like and find very fast. And it can be a familiar UI to mayer users (note that I am not claiming that one is better than the other.. Thats subjective).

Re:Rocks, as in "there's a rock in your shoe"?

[argent]

Intuitive for *anything* is [...]

I didn't say one word about "intuitive".

Blender's user interface is just bad. It's needlessly complex. It's not simply subjective... people who are familiar with all kinds of user interfaces have similar opinions of Blender's. Hell, I've never used Maya, and the three view non-overlapping windows model in Blender appears superficially similar to the first 3d program I used - Sculpt 3d - so if it was a matter of familiarity I'd be right on top of it, wouldn't I?

And customizable user interfaces has been a non-solution to bad user interfaces forever. That way lies Emacs.

Yeh, I can see Blender becoming the Emacs of 3d UIs. Same complete lack of attention to the basic UI design, same "stockholm syndrome" defensiveness from the fans. Same bad emulations of other programs that completely fail to hide the unnecessary complexity.

Re:Rocks, as in "there's a rock in your shoe"?

[delt0r]

Good verse bad is subjective. There is also a lot of people with a different option than yourself.

Re:Rocks, as in "there's a rock in your shoe"?

[argent]

I wonder how many other people who argue that it doesn't matter whether a program is "good" or "bad" can't spell even with a spelling checker?

Download?

[mwvdlee]

So, where can we download this tool and try it for ourselves?

openSUSE on X60 Tablet MultiTouch

[cflange]

I recommend trying OpenSUSE on your tablet. I am not sure about Fujitsu, but my Lenovo X60 Tablet has been working fine with digital pen since openSUSE 10.3. The openSUSE guys have been very active in developing new support for tablets. Check out this page: http://en.opensuse.org/TabletPCs#Configuring_Your_Tablet_Device

With the beta version of openSUSE 11.1, which I am now testing, I can get even the touch screen mode of my MultiTouch tablet working. Add to it the 3D desktop effects and it blows everyone out the window!

Re:openSUSE on X60 Tablet MultiTouch

[Nightspirit]

Thanks, I'll try that.

Re:So this isn't actually useful

[Cathoderoytube]

Your reply was so astute I completely fell back in my chair and cracked my skull on the table behind me, then had to go to the hospital and get 23 stitches.

Seriously. It's all the trend of the people who do all the programming having no idea what the hell people use their software for and just program things they think it's cool but artists have no bloody use for. People who program 3D especially are as unreceptive to change requests as they come. Hence my disgust at this new useless software.

Re:So this isn't actually useful

[xenocide2]

This is research. I don't what planet you're on, but where I come from the point is to push the frontier of what is possible, less than to push the frontier of what's useful. Less so would be "what's useful in Maya." Fixing bugs is not science. It's very boring engineering work and not something we should be allocating our best and brightest researchers for, to the benefit of a private company. Be angry, if you must, but be angry at Maya, not the people who don't own, don't control, and have no access to Maya or it's software.

Re:So this isn't actually useful

[robthebloke]

The big problem is that there is very little competition left in the 3D industry. Maya, Max and MotionBuilder are all owned by Autodesk - unfortunately the biggest competitor to Maya, is 3ds Max, which isn't really competition at all...

XSI/modo/lightwave/cinema4D/rhino are all alternative products, but unfortunately their market share is too small to make too much difference.

Then lets talk about innovations in the 3D graphics world - since the geforce 1 was introduced, new and original research has largely been replaced with papers that put 10+ year old techniques onto the GPU. The GP has a very valid point here, there is very little research being done that actually provides new and useful improvements to the users actual workflow.

The problems mentioned by Cathoderaytube that he finds in Maya, are all present in the other packages as well. These are not 'bugs', they are mathematical flaws that need some pretty dedicated R&D to come up with new solutions. If you are about to embark on a Phd in 3D graphics, go and talk to an animator for a day, and you'll very quickly find yourself with 50+ research topics to investigate. I think it's safe to say that there wouldn't be a single GPU topic in that list.... but if you implement just 1 for your Phd, you'll have a few hundred thousand users saying thankyou.