Jeremie, I'm confused. You guys rig in Maya but animate in Softimage? On Sun, May 20, 2018 at 5:46 PM, Sebastien Sterling < [email protected]> wrote:
> That is super generous, wish more people would share comprehensive artist > friendly stuff > > On Sat 19 May 2018, 20:43 Jeremie Passerin, <[email protected]> wrote: > >> Anytime. Blur is not keeping much secret. We believe the talent of our >> crew and constant search for innovation is what keeping ahead of the game. >> Our blend shape solver is open source and works in Softimage and Maya. >> >> https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_blurstudio_Simplex&d=DwIFaQ&c=76Q6Tcqc-t2x0ciWn7KFdCiqt6IQ7a_IF9uzNzd_2pA&r=GmX_32eCLYPFLJ529RohsPjjNVwo9P0jVMsrMw7PFsA&m=A6i0Eh_b5LRtfYvLhTOjMwfLchpggiTVI59LVQxzcXg&s=CM_1MDGYJ2yHGd80OJ_aI_Fz0-5RrxoO36Ss1PTrDFQ&e= >> <https://u7507473.ct.sendgrid.net/wf/click?upn=5SmYwFIJXHmC5X9wAP0G6mg4oLGBuQENbeDkYXezg3m6vjHxJcC6rUMd8QE2MtqzS3nHFJpHH1mihN2vNWVF73yH4QzPTUaj-2BU-2FV4xCvLLTnZ7suFMzmFdGmjVRnYdJAxVY-2F5M380PkfQ8TNK8ovM5rxNeeTCOnkRGaT-2F-2Fqta38pWvWMaql23pwdaAWmes-2Fu-2FwoONKissOPCP2dnxYlCF6yfmwOfjSUg-2BFWpyDZ9PlLILY2VTJIavQv1WJneE5K5CMw7TU-2FZfZsCIiIDMFnc3JXUbPqO8l-2BprbR1JcC9ABVTDHN9oP2B9SI37dfH9LcCeytAli4HqahpAD5CInhd0l3sl9vmaRK0p-2Bcgs6QvcfJV-2BBT62tQgwG7ZZsXO-2BYVpHR9PpM9GSrxwsipz5EomeIpsF8Sc1tqfvhjfFzrl8j4nIJPENMVVnXPq7WwD3dGeoF5mdZLCYRLYdrvBUCF3CuvqazIJBlDsF9wiYqY8ZgdGjb3tsY6maC-2BZUnRIqsY5mKsNAAWmbRcq-2BKgEIa4Jxk9x3fzcFFqfuoZ-2BQGxzRlf-2BBfgE9Fixld7GNaX9W51DnELQm9EY33gViVPT5bjEp0qq5Y6SFlmvOM9T0ve5rvztuFRLhcoUR02UzfHOWHPjpQafnfofoXU4fglCJ-2BjLA5mqps8-2BUpoBmw1rR85TE1Iydif0P2neYQPBJ8WmDVFq1aPlGnlIewsNu9B0PFt1y6eNvEddsZZwXzB9werX3As2YcKwsZyvvHJIz8HAwL-2FG9uoaCmLZ8fn5u0p2UrUv4taf4RGfeyxFgUZi9bfATeSxQEpa1x7UI28llS7i-2BbTpaGoMK6RhVqgmNEWbsxSO0njlhu5Q8cblFo2VS2eWMkKAnpsGLX8YljcjKEownSiYVY9Kwb0m5YQPDqZT4NgYWqyrVwQf0vHrfLCkjxfZHULIdYssLP0pLgvlgGoKlZuEZ1PccJsBxZynIY-2FXN8WB1oX0kcDNXunuqllM8K5DwkpKasCgUN92IQdMNyU-2FOWQo0lF-2FX9xLS8g-2FR6cCkiK-2BmnSVdYxk9DtvP4Vc0CZyhxml3CwINe3gcwVo-2FCGM-2B6JMVP1H3HubnFjixVdDPigpcEhjjtITfqE-2FiOFqyqr-2Fzq2PMIDQvweoJMOdWLvfJiZIGJM2jqXDsBtgQXBMFmxapfhP-2FZu7uyCIM0JtuYqpBnvWEH994d0uHUQK4FTGNRsITAFYWLz3cJfl96SebXYKitVVksMGnr1S0-2FK4zzNxedEb4CJub4r67vtGuhYGkVTmD6-2BKg45flJJ6Tt0tl17bSA-3D-3D_wAli21vZ0ansRLHjQSxwcqGxGljxven5UMtKBZQLjG1-2FviMnbkVLZef58JWfdr1OqwQ3OSEg-2FKbu43zv5T3DDwQgQJgBjPYQ6lT08yQen29FqOiXHOxBMEPr9MjG22jrbl5E8OAZ9abOCt1FHVhKjrMilQF51w4dXLJPX9vTmIY9ZEoIu3ZuIp-2BVk4A8jia0EZ-2BjUX2d-2BjdwbeE8ZNpL-2FRfizDuQ5PjCO4uEGYV2wwk-3D> >> >> On Sat, May 19, 2018, 8:38 AM Sebastien Sterling < >> [email protected]> wrote: >> >>> Well, thanks for your time in answering these questions Jeremie. Best of >>> luck to you guys at blur. >>> >>> On Sat 19 May 2018, 15:49 Jeremie Passerin, <[email protected]> wrote: >>> >>>> In our case it wouldn't be possible to model the 500 shapes I mentioned >>>> on each character for each project considering our tight deadline. >>>> We've developed some techniques to create a standard set of blend >>>> shapes that we can transfer from one character to the other. >>>> The transfer is usually good enough for background characters and >>>> simple expressions but we have to clean the shapes for hero characters.That >>>> means you always start with a base and you clean the deformations and match >>>> the references. >>>> >>>> The sculpting of the shapes is done in Maya. Because you already have a >>>> base, we usually don't need to use deformers, it's all hand sculpted. We >>>> used deformers to create the first set of shapes though. >>>> The mesh is dense but Maya sculpting tools are pretty good. We still >>>> animate in Softimage so we have tools to transfer over the shapes and build >>>> the facial rig. >>>> >>>> Note that the facial modeler is not necessarily the facial rigger. >>>> Those are different skills and not everyone can do both. Our best facial >>>> modeler has studied facial anatomy for years and that's pretty much all it >>>> does. Even our top modelers can't do what he does, and I certainly can't do >>>> any of that. >>>> >>>> >>>> On May 18, 2018 11:23 PM, "Sebastien Sterling" < >>>> [email protected]> wrote: >>>> >>>> To jeremie. >>>> >>>> When someone comes to work for you, do you train them in-house on your >>>> methods ? Do you have an in-house training Manuel to such effect detailing >>>> the various expressions and breakdowns e.g lip curled in/out mouth corners >>>> up/down... and how to make them using deformer as supposed to just manually >>>> moving point, which doesnt reaฤบy work if you are working with very dence >>>> meshes and trying to keep your blending smooth. >>>> >>>> I'm sure you have proprietary tech you don't want to share with other >>>> studios, and this is not what I am asking for. I just want to know how >>>> shapes are built using deformers and the breakdown of which shapes are >>>> used. >>>> >>>> E.g. when you are making the eyes open, you paint the entire closed eye >>>> cluster, then place the pivot in the center of the eye and rotate up and >>>> down, which will eventually give you your eyes open and your lid zip.... >>>> >>>> I'm not sure if I'm asking to much and if these are not things you are >>>> allowed share. >>>> >>>> On Sat 19 May 2018, 07:07 Jeremie Passerin, <[email protected]> >>>> wrote: >>>> >>>>> BCS allows for two important things if you build your facial with >>>>> morphs. >>>>> Non linear interpolation and combo corrective. The second one being by >>>>> far the most important. >>>>> This is our approach at blur. We're not using bcs, we have our custom >>>>> tools for that. It's also how they are doing it at weta last i heard (We >>>>> work we two ex-weta facial modelers). >>>>> Our setup has about a hundred primary shape and more than 500 hundreds >>>>> with all the combos. >>>>> Now how it's build, I'm not trying to keep it a secret but it's >>>>> complicated. >>>>> Most studios are trying to approach them as FACS. It's theoretically >>>>> giving you all the possibilities of the human face but it's not super >>>>> artist friendly and also we often need to exaggerate/cheat what human can >>>>> do. So it's often not enough. >>>>> We also have a ton of tweak controllers on top of that. Not to mention >>>>> special deformers such as cornea bulge and lip zipper. >>>>> Monster in Paris was stylized and the setup wasnt nearly as complex at >>>>> what blur is doing today. >>>>> Disney has a different approach based on curves deformers and >>>>> corrective psd last I heard. >>>>> If you have more precise questions. I'd be happy to answer with my >>>>> knowledge of the question. >>>>> >>>>> On Fri, May 18, 2018, 4:20 PM Sebastien Sterling < >>>>> [email protected]> wrote: >>>>> >>>>>> At the mo I'm looking into the most obvious methods; the blend >>>>>> shape/bcs method. >>>>>> >>>>>> Creating an initial set of clusters on the mouth corners, brows etc... >>>>>> >>>>>> Then using them to extract your shapes, and a second time at the end >>>>>> as a secondary layer of control. >>>>>> >>>>>> It would be nice to find some breakdown, of how this stuff is >>>>>> supposed to work, what shapes are required etc... >>>>>> >>>>>> It's a difficult one to explain, as I know most places guard this >>>>>> stuff like gold dust. >>>>>> >>>>>> To Jeremie Passerin. >>>>>> >>>>>> I'm sure you'd have a lot to teach inspite of your current work not >>>>>> being feature film Jeremie ๐๐ >>>>>> >>>>>> To Olivier >>>>>> >>>>>> You must have known David Galente as well ๐ >>>>>> >>>>>> On Sat 19 May 2018, 00:56 Sebastien Sterling, < >>>>>> [email protected]> wrote: >>>>>> >>>>>>> Hey people, sorry for the late response. >>>>>>> >>>>>>> Basically I'm looking for any decent information on facial rigging. >>>>>>> Am trying to level up to showcase some characters, I'd like to do so in >>>>>>> a >>>>>>> way that could potentially interest some employers. >>>>>>> >>>>>>> Facial rigging is pretty obscure field of study, I have most of the >>>>>>> available materials, and they are not cutting it. >>>>>>> >>>>>>> So yes hippydrone and yes Fahrenheit facial rigging. >>>>>>> >>>>>>> I'm interested in what people actually use in production Now, not so >>>>>>> much what was cool in 2005 :p >>>>>>> >>>>>>> I was told that a monster in Paris had a pretty cool work flow from >>>>>>> some people who worked there, but I don't have too many leads, this >>>>>>> would >>>>>>> be in Maya, but what I'm after is the philosophy and methodology of >>>>>>> facial >>>>>>> rigging. >>>>>>> >>>>>>> Sorry if this is vague. >>>>>>> >>>>>>> >>>>>>> ------ >>>>>>> Softimage Mailing List. >>>>>>> To unsubscribe, send a mail to softimage-request@listproc. >>>>>>> autodesk.com with "unsubscribe" in the subject, and reply to >>>>>>> confirm. >>>>>> >>>>>> ------ >>>>>> Softimage Mailing List. >>>>>> To unsubscribe, send a mail to softimage-request@listproc. >>>>>> autodesk.com with "unsubscribe" in the subject, and reply to confirm. >>>>> >>>>> ------ >>>>> Softimage Mailing List. >>>>> To unsubscribe, send a mail to [email protected] >>>>> with "unsubscribe" in the subject, and reply to confirm. >>>> >>>> ------ >>>> Softimage Mailing List. >>>> To unsubscribe, send a mail to [email protected] >>>> with "unsubscribe" in the subject, and reply to confirm. >>>> >>>> >>>> ------ >>>> Softimage Mailing List. >>>> To unsubscribe, send a mail to [email protected] >>>> with "unsubscribe" in the subject, and reply to confirm. >>> >>> ------ >>> Softimage Mailing List. >>> To unsubscribe, send a mail to [email protected] >>> with "unsubscribe" in the subject, and reply to confirm. >> >> ------ >> Softimage Mailing List. >> To unsubscribe, send a mail to [email protected] >> with "unsubscribe" in the subject, and reply to confirm. > > > ------ > Softimage Mailing List. > To unsubscribe, send a mail to [email protected] > with "unsubscribe" in the subject, and reply to confirm. >
------ Softimage Mailing List. To unsubscribe, send a mail to [email protected] with "unsubscribe" in the subject, and reply to confirm.
