Tagarela is free and open-source editor and runtime solution to create facial animations to your 3d characters.
The main focus of the project is to create an easy-to-learn/easy-to-implement system, trough a funny and small keyframe editor.
The system uses vertex manipulation to create the interpolation between blend shapes. Shapes can be created in your favorite 3d application, like Autodesk 3dsMax/Maya/XSI or Blender.
Using the built-in WYSIWYG Animation Editor, you can edit your animations by adding and dragging keyframes, setting the morph shape sliders to create new expression, while you see the transformations and hear the audio in the exact keyframe position. All was prepared to make the editing fast and funny.
The current version supports 2 kinds of animations:
Using an audio file: the animation will follow the audio current play-time to prevent latency.
Custom animation length: will play based on unity timer, and don´t play an audio.
Live Demo
Click here to check the real-time demo application (Unity web player)
Features
Built-in keyframe editor
Audio waveform image preview
Real time preview in Unity scene editor
Shift+drag to easily clone a keyframe
Simple call Play(“yourfile”) to run your animation
Audio animations are sync to audio.time, to avoid latency
Download
Click here to download links and more info. Source-code (GNU GPLv2) and Compiled DLL´s (as an alternative to GPLv2) available.
About the project
I have been working on this project in my free time, during this last year (2012). Use it at your own risk.
Let me know if your are using this tool in your project, i´ll be very glad.
Hey Rodrigo, nice work! This is awesome. I’ve been testing it with my 3dmax model but i have encountered a problem. The problem is when I am going to add the different Blend Shapes, Tagarela shows me an error “Blend Shape Vertex Count must have the same that the Main Object”. It is very strange because the morpher that I have made in 3dmax is made from the same main object and the vertex count should be the same.
@jdth: hello, i´m glad you like it. There are some simple solutions to fix this:
– on Inspector Panel of your 3d model (selected in the Project Panel), disable the
option “Optmize Mesh”;
– place all models inside one .fbx instead of one fbx per model;
– and under the option “Normals Tangents” (located in the same inspector panel)
change the value of the option “Normals” from “Import” to “Calculate”, and set the
“Smoothing Angle” slider to 180.
Take a look at the Documentation.pdf inside the .unitypackage, there is a datailed description with an image, if you have problems to follow the paths above!
Rodrigo
I have question regarding placing all models inside one .fbx. How do you do it in Blender? I have made lipsync animation with Papagayo and MHX lipsync, but I don’t know what to do in blender to get the blend shapes inside unity like in your demo.
I mean how do you get separate pig heads (blend shapes) inside the fbx file;
I have tried to locate tutorial for how to do this, and have been experimenting my eyes out with blender but to no avail. I’m just so new to blender and unity that I need a little help to start using this awesome tool for which I’m so thankful!
I’m wondering if its possible to do the same thing but with single frame animations. Considering UMA for example, and how its all connected with Mecanim to provide a generic base for any model it would be really awesome to see this fantastic tool’s front end working with single frame animation positions of a skeleton instead of fbx models.
I didnt even know this existed, it looks great! Thing is, Unity has blendshapes in the box now (since 4.3.3 or so i think), so with a little fiddling you can get straight to your blendshapes that would be in your character’s main fbx. There’s a system on the asset store for automating lip sync, much fewer blendshapes needed, not as accurate but not so time consuming. However how do you automate facial expressions? Not easy at all i’d imagine, this looks great for keyframing expressions, who volunteers to get it working with the new blendshapes?
I’m bookmarking this now, if noones done this by the time I feel i could use it, i’ll have a go
I suppose that may answer your question, LaneFox. What you just asked was pretty much this minus the blendshape import, just needs tieing up with Unity’s blendshape import. You can save a character’s morphs/blendshapes and skinning in a single fbx file.
Actually i just checked the comments on the website, is it UMA that specifically does not have blendshapes? I can understand why, but i suppose for those using blendshapes, its fine
Well, we’ve been talking about it on the Skype channel. I think Eli is currently working on something so in a few weeks perhaps we’ll have something to do that with.
I asked Rodrigo(Pegorari) about it on twitter and since Tagarela didn’t really get very popular he’s not motivated to upgrade it.
Is the code still functional without any errors or questionable warnings? Havent tried it yet, but for blendshapes it’s almost a downgrade, a bunch of the clever stuff in there doesn’t need to be there, or at least used unless you have a specific need outside unity’s implementation of blendshapes. Saying I don’t have time because of other products is a bit weak consider ive spent the day wasting time instead of working on my project but no point me going into it just yet, also i’m dim with programming, someone sharper than me at coding (pun not intended although it should have been) should have this up to that and even using the keyframe editor without audio, say a space of time for events or linked up to whatever mecanim’s doing, or collection of blend weights for in-context random stuff
Like I say though there’s a recently released very reasonably priced asset store thing for decent automated lip sync http://forum.unity3d.com/threads/242135-SALSA-with-RandomEyes-1-0-0-RELEASED, this however seems just the ticket for expression animation though right now (Or even adding extra movement to whatever salsa is doing). Not even just lip sync, blendshapes are pretty flexible (Pun kind of intended that time)
I’m currently using Daz stuff while UMA matures a little. You can save umpteen body, face etc blendshapes (although obviously keep it practical) along with the clothes and the blendshapes to match the figure with Daz’s FBX exporter, along with rigging that easily matches a humanoid avatar in mecanim and there’s your bajillion configurations. It’s a lot heavier than UMA, and obviously a bunch of money (although nothing like employing someone to do that), i’d like UMA to have some kind of functionality like that, but it’s currently not got the visual variety, particularly the amount of authored easily available content, for me to go that way right now. I’m one person trying to keep their work interesting, while ostensibly an artist, i would never get anything done trying to create enough art to get UMA enough content to suit me at the moment, compared to the other option, right now. If i’m wrong certainly tell me, i’ve not been in the community skype, i’d probably be a bit behind on what regulars consider status quo to the point of appearing dumb
So in that sense i’m very glad (you’d have to tell me who Eli is) is doing great work with this, getting it working with UMA is great, is he planning to include unity’s blendshapes as another source of targets? Tell him i’d very much appreciate it!
Eli is a C# coder and I did try to work with UMA and there are excellent coders and contributors on UMA. I even wrote about 3 or 4 thousand lines of code that I haven’t finished yet (because of something like ADHD , maybe someday) . I looks at a link Unlogick had of a web player Eli made of his work and Eli is basically going to make it were UMA can do the full range of face acting so impressive indeed.
Unlogick Fernando last I knew are adding quadrupeds and they already look good. Lucky animals don’t need to change clothing.
For Eli’s work it will use bones and I’m sure he’ll probably create a list of expressions you can call in sequence to read the Gettysburg address or react to game events. It really is good.
It’s true there is not visual variety yet.
Oh, I don’t log onto Skype because of the ADHD but I check the UMA thread in the Asset Store portion of the forum occasionally.
Got pals with that, its harsh, so im sorry to hear. Seen the animal work they’ve done, it’s sterling, and i’m very glad its getting a lot of support in general, i’d like to be one of you folks who are fully behind it one day (As in using it consistently throughout projects, or writing things for it), i’m sure i probably will
, maybe someday) . I looks at a link Unlogick had of a web player Eli made of his work and Eli is basically going to make it were UMA can do the full range of face acting so impressive indeed.