Alright FINALLY what I've been looking for, with hands and them moving naturally! Thank you! Tho.. there are many things I'm unfamiliar with made me lose you🤧😂 so could you please tell which program you've used for this model? And how did you make it? Like did you draw them in separate layers like ones for 2D softwares?
Great video, great demonstrations! One other good point of using an Android or iPhone for your face tracking is that it offloads a bunch of processing from your computer. I've got a sack of potatoes for my stream PC and since picking up iPhone tracking, I find I have less strain on my CPU to go with my fancy, fancy expressions!
I hate being so confused, so I'm sorry to ask, but I'm trying to understand what programs I need specifically in order to make the best content with an iphone. I don't know what apps I need. I promise, I looked at the description!! I'm just still....very confused.
i'll do my best to help! this video is mostly meant to help show the difference in face tracking, but here's my recommendations: 1. you need a 3D model for my method. live2D is a bit different! 2. your 3D model needs what are called blendshapes to take advantage of the nice tracking on your iphone (shapes... that blend together... make sense?) for a vroid, you need to export it as a VRM and then follow the hanatool tutorial i provided in the description. lady aska has a fantastic one, but it's not the only one. this will apply 52 different "blendshapes" to your vrm model that the iphone will connect to (more detail about this in the linked *-- +tutorial). - at this point you have a model that CAN use iphone tracking... so how do you use it? 3. ifacemocap is an iphone app that uses your iphone camera to collect face tracking data. you'll need to install it and follow the included instructions. 4. vseeface takes your model and the instructions from ifacemocap and actually applies the face tracking to your model. you need to install and set up vseeface so that it listens to ifacemocap. 5. test your model to make sure it behaves how you expect. you may have to go back to a previous step to tweak things if they're not correct! - now you have a model following your faces movements... but how do you stream? 6. connect vseeface to obs. vseeface includes instructions on how to capture your model in obs, but there are also some great tutorials on youtube. 7. test obs by recording a video of your model on top of your background. - if that works as expected, you're ready to set obs up to go live! i hope this is helpful! there are a bunch of tutorials on each step of the process, but i hope this gives you a general guide to follow. :)
Mah bad, I am unfamiliar with "Advanced blend-shapes". It's kinda my fault because I only been following tutorials about PNGtubers so perhaps this is a whole new technique/technology? I have some vague understanding of animation with Adobe Character Animator and also seen what Live2D is doing for people but honestly, I am a complete luddite with these terminologies and techniques! >_< Please help!
"blendshapes" are a bit like animation key frames for 3D models - you pose the model and the computer can combine multiple poses into a whole new one. for example, you might pose the model with one eye closed, and also pose it smiling. then you can "blend" those two "shapes" into a new one where the model is winking and smiling. for the high quality motion capture from your iphone to work on your 3D model, you need to save 52 different "blendshape" poses to your model (collectively called "perfectsync"). these include things like showing teeth, squinting eyes, closing the mouth, etc. without these your model can't be articulated precisely enough to take advantage of all the information your motion capture is sending to your PC.
Me: * Happily bouncing with excitement to learn new things to be able to work on * Sure can't wait to take my VTubing with consistent low live stream view counts to the next level, Oh Bo- Moth: By the way, if all this still gets low views- Me: Thats fine, to be honest the work behind it interests me and I do want to at least attempt to make my content better.. Moth: Then you may want to improve THE CONTENT. Me: OOOF!! Why Moth, why? I mean, you are only trying to provide as much help, yes. And you aren't wrong, improvement is literally the point on why I am here, but why did you need to roast us that much? XD JK of course lol. Only messing around X)
ahh my friend! i hope i didn't across too harsh... i think sometimes people get worried and rush out to buy more expensive equipment, and i wanted to reassure folks that the expensive option isn't always necessary... maybe i was a little too keen on that though lol ultimately i feel your excitement on making those small upgrades... if you love to see things slowly get better and working on improving your model is your thing, you can for sure see the difference. to be fair that's exactly what i did, and i'm no xqc lmao i'm sure your content is fantastic my dear, and as you said it can only get better!
On a not joking note, I watched a bit on repeat to see the differences between Basic Blend Shapes with Cam and Hana Blend Shapes with Cam. And honestly I see not much difference lol. Then again, I am imagining the Blend Shapes on my own Model, which already has decently customized Blend Shapes through VRoid Studio to avoid the overexpression problems they have from the start along with some funny faces I threw in there. I am sorta wondering if the Hana Tool would still be suggested even after some fiddling about, Specifically for Webcam only.
@@misterkillhappy so the difference between the "fancy" blendshapes and the "simple" blendshapes is quite large the hana app allows a set expressions to work - "happy", "sad", etc the hana tool includes 52 "perfectsync" blendshapes - eyebrows up in the middle, mouth pulled to the side, left eye wider in the inner corner, things like that the webcam limits a lot of that movement - you can't see it because the camera can't pick up on it and transmit that data to vseeface. that being said i would still recommend including the more advanced blendshapes for 3 reasons: 1. hana tool and hana app cost the same. if you're going to pick one, pick the tool 2. the fancy blendshapes future proof your model. it'll be ready for you if / when you upgrade to better tracking, and will allow you to use apps like VRM posing desktop to articulate your models face for other things like graphics / "photoshoots" 3. learning how to edit the blendshapes on your model opens up a bunch of really handy skills... for example, if you don't buy an iPhone for tracking you can add canned animations (i have an idle body animation i use which i think adds a lot), extra textures like hearts, sweat drops, sparkles, blendshapes for clothing items or accessories, etc if your model moves in a more simple way anyway, these extra animations and graphics can give a really charming anime-ish effect which is way more noticeable to the viewer than whether or not you can tug your upper lip up (generally speaking, in my opinion) i hope this is helpful! there are a lot of great tutorials on how to add those animations and stuff so have a look around :)
Alright FINALLY what I've been looking for, with hands and them moving naturally! Thank you!
Tho.. there are many things I'm unfamiliar with made me lose you🤧😂
so could you please tell which program you've used for this model?
And how did you make it? Like did you draw them in separate layers like ones for 2D softwares?
Great video, great demonstrations! One other good point of using an Android or iPhone for your face tracking is that it offloads a bunch of processing from your computer. I've got a sack of potatoes for my stream PC and since picking up iPhone tracking, I find I have less strain on my CPU to go with my fancy, fancy expressions!
yes this is a really good point! the phone handles a ton of processing, which lets your PC focus on other stuff!
Yeah this is why I even off load OBS to a second PC. The gaming PC is really a notable potato among potatoes...
What did you use to create your avatar my wife is wanting to start vtubing a we are just looking for tips
I hate being so confused, so I'm sorry to ask, but I'm trying to understand what programs I need specifically in order to make the best content with an iphone. I don't know what apps I need. I promise, I looked at the description!! I'm just still....very confused.
i'll do my best to help! this video is mostly meant to help show the difference in face tracking, but here's my recommendations:
1. you need a 3D model for my method. live2D is a bit different!
2. your 3D model needs what are called blendshapes to take advantage of the nice tracking on your iphone (shapes... that blend together... make sense?)
for a vroid, you need to export it as a VRM and then follow the hanatool tutorial i provided in the description. lady aska has a fantastic one, but it's not the only one. this will apply 52 different "blendshapes" to your vrm model that the iphone will connect to (more detail about this in the linked *--
+tutorial).
- at this point you have a model that CAN use iphone tracking... so how do you use it?
3. ifacemocap is an iphone app that uses your iphone camera to collect face tracking data. you'll need to install it and follow the included instructions.
4. vseeface takes your model and the instructions from ifacemocap and actually applies the face tracking to your model. you need to install and set up vseeface so that it listens to ifacemocap.
5. test your model to make sure it behaves how you expect. you may have to go back to a previous step to tweak things if they're not correct!
- now you have a model following your faces movements... but how do you stream?
6. connect vseeface to obs. vseeface includes instructions on how to capture your model in obs, but there are also some great tutorials on youtube.
7. test obs by recording a video of your model on top of your background.
- if that works as expected, you're ready to set obs up to go live!
i hope this is helpful! there are a bunch of tutorials on each step of the process, but i hope this gives you a general guide to follow. :)
Mah bad, I am unfamiliar with "Advanced blend-shapes". It's kinda my fault because I only been following tutorials about PNGtubers so perhaps this is a whole new technique/technology?
I have some vague understanding of animation with Adobe Character Animator and also seen what Live2D is doing for people but honestly, I am a complete luddite with these terminologies and techniques! >_< Please help!
"blendshapes" are a bit like animation key frames for 3D models - you pose the model and the computer can combine multiple poses into a whole new one.
for example, you might pose the model with one eye closed, and also pose it smiling. then you can "blend" those two "shapes" into a new one where the model is winking and smiling.
for the high quality motion capture from your iphone to work on your 3D model, you need to save 52 different "blendshape" poses to your model (collectively called "perfectsync"). these include things like showing teeth, squinting eyes, closing the mouth, etc. without these your model can't be articulated precisely enough to take advantage of all the information your motion capture is sending to your PC.
Me: * Happily bouncing with excitement to learn new things to be able to work on * Sure can't wait to take my VTubing with consistent low live stream view counts to the next level, Oh Bo-
Moth: By the way, if all this still gets low views-
Me: Thats fine, to be honest the work behind it interests me and I do want to at least attempt to make my content better..
Moth: Then you may want to improve THE CONTENT.
Me: OOOF!! Why Moth, why? I mean, you are only trying to provide as much help, yes. And you aren't wrong, improvement is literally the point on why I am here, but why did you need to roast us that much? XD
JK of course lol. Only messing around X)
ahh my friend! i hope i didn't across too harsh... i think sometimes people get worried and rush out to buy more expensive equipment, and i wanted to reassure folks that the expensive option isn't always necessary... maybe i was a little too keen on that though lol
ultimately i feel your excitement on making those small upgrades... if you love to see things slowly get better and working on improving your model is your thing, you can for sure see the difference. to be fair that's exactly what i did, and i'm no xqc lmao
i'm sure your content is fantastic my dear, and as you said it can only get better!
On a not joking note, I watched a bit on repeat to see the differences between Basic Blend Shapes with Cam and Hana Blend Shapes with Cam. And honestly I see not much difference lol. Then again, I am imagining the Blend Shapes on my own Model, which already has decently customized Blend Shapes through VRoid Studio to avoid the overexpression problems they have from the start along with some funny faces I threw in there. I am sorta wondering if the Hana Tool would still be suggested even after some fiddling about, Specifically for Webcam only.
@@misterkillhappy so the difference between the "fancy" blendshapes and the "simple" blendshapes is quite large
the hana app allows a set expressions to work - "happy", "sad", etc
the hana tool includes 52 "perfectsync" blendshapes - eyebrows up in the middle, mouth pulled to the side, left eye wider in the inner corner, things like that
the webcam limits a lot of that movement - you can't see it because the camera can't pick up on it and transmit that data to vseeface. that being said i would still recommend including the more advanced blendshapes for 3 reasons:
1. hana tool and hana app cost the same. if you're going to pick one, pick the tool
2. the fancy blendshapes future proof your model. it'll be ready for you if / when you upgrade to better tracking, and will allow you to use apps like VRM posing desktop to articulate your models face for other things like graphics / "photoshoots"
3. learning how to edit the blendshapes on your model opens up a bunch of really handy skills... for example, if you don't buy an iPhone for tracking you can add canned animations (i have an idle body animation i use which i think adds a lot), extra textures like hearts, sweat drops, sparkles, blendshapes for clothing items or accessories, etc
if your model moves in a more simple way anyway, these extra animations and graphics can give a really charming anime-ish effect which is way more noticeable to the viewer than whether or not you can tug your upper lip up (generally speaking, in my opinion)
i hope this is helpful! there are a lot of great tutorials on how to add those animations and stuff so have a look around :)