正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

Face angles rigged with a depth map! It's so lightweight, AE isn't even lagging 😭🥀 #aftereffects

Ezboy | VGen

6,077 subscribers

127,002 次观看 • 10 天前 •via X (Twitter)

艺术科学技术 #aftereffects

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

the live keeps lagging so mingyu kept posing prettily while talking so he doesn't make a funny expression when it pauses 😭😭 #민규 #세븐틴 he's such a comedian 😭 but he looks pretty in all angles anyway!!!

the live keeps lagging so mingyu kept posing prettily while talking so he doesn't make a funny expression when it pauses 😭😭 #민규 #세븐틴 he's such a comedian 😭 but he looks pretty in all angles anyway!!!

shanie🍥 ggyu

43,614 次观看 • 7 个月前

First this now this complete robbery of Lomachenko 😭😭 Boxing as a sport so rigged and they not even hiding it now💀

First this now this complete robbery of Lomachenko 😭😭 Boxing as a sport so rigged and they not even hiding it now💀

ᕋOO⅃

170,728 次观看 • 3 年前

I'm crying at how Taehyung jokes around and plays with everyone he's so funny 😭😭😭 👤:Isn't this a hat? Isn't it yours, Taehyung? 🐻: I don't think it's mine 👤: It's not? Alright then, you can go~ Thank you for your hard work 🐻: Oh, wait a second! 🐻:It might actually be mine... Ah, no, no, it's not!😭

I'm crying at how Taehyung jokes around and plays with everyone he's so funny 😭😭😭 👤:Isn't this a hat? Isn't it yours, Taehyung? 🐻: I don't think it's mine 👤: It's not? Alright then, you can go~ Thank you for your hard work 🐻: Oh, wait a second! 🐻:It might actually be mine... Ah, no, no, it's not!😭

非 𝐇𝐈𝐁𝐓𝐀𝐄 ꪜ

41,134 次观看 • 2 个月前

KILLER T just delivered a legendary performance. 🔥 This isn't just a song, it's the ANTHEM of the year. No cap, it's impossible to even hate I'm genuinely impressed. 😭🙌👍

KILLER T just delivered a legendary performance. 🔥 This isn't just a song, it's the ANTHEM of the year. No cap, it's impossible to even hate I'm genuinely impressed. 😭🙌👍

J E T H R O 🇿🇼

25,558 次观看 • 10 个月前

The way nothing ever embarrasses Nicki “Even with that face I still look cute.” She’s such a kii 😭😭😭

The way nothing ever embarrasses Nicki “Even with that face I still look cute.” She’s such a kii 😭😭😭

She’s Fishy ➐

29,872 次观看 • 2 个月前

Security Guard gets his face stretched with a fish hook by a crackhead customer. How does that even happen?😭😭

Security Guard gets his face stretched with a fish hook by a crackhead customer. How does that even happen?😭😭

STREET MEDIA TV

180,925 次观看 • 3 年前

SANI BEING SO AMUSED AND JUST WATCHING YUNHO WITH A SMILE ON HIS FACE 😭😭

SANI BEING SO AMUSED AND JUST WATCHING YUNHO WITH A SMILE ON HIS FACE 😭😭

ㅁㄹ

30,741 次观看 • 5 个月前

THE DEPTH MAP TRICK THAT FIXED DANCE ACCURACY IN SEEDANCE 2.0 Feed the model a video of someone dancing and it tries to interpret everything- the person, the clothes, the lighting, the room, and somewhere in there, the movement. Feed it a depth map and there's nothing left to interpret but the motion. Most creators trying to transfer a dance to a character reference the source footage directly, then wonder why the choreography drifts. The problem isn't the model - it's that you handed it ten variables when you only wanted one. Here's the workflow 1. Lock the character reference in GPT Image 2 first -face, build, costume, so identity holds independently of whatever motion gets applied to it 2. Convert the source dance footage into a depth map instead of using the raw video -this strips out the original performer's appearance, clothing, and environment entirely 3. Feed the depth map as the motion reference and the character sheet as the identity reference- two separate inputs doing two separate jobs, not one input trying to do both 5. Let the depth map carry only spatial movement -the model receives body position and momentum with no competing information about who's moving or what they look like 6. Keep the character and motion inputs isolated throughout - the moment you mix appearance data into the motion reference, the model starts negotiating between two identities Why this works • Raw footage passes the model everything at once- performer, wardrobe, room, lighting -and the choreography competes with all of it for attention • A depth map is pure spatial information, so the only thing left to transfer is movement • Separating identity from motion means the character can stay locked while the dance stays accurate - normally you're trading one for the other • The accuracy gain isn't the model getting better, it's the model getting fewer decisions to make Use cases: ⁃ Dance and choreography transfer onto original characters ⁃ Motion capture-style workflows without motion capture ⁃ Any sequence where a specific movement needs to survive intact ⁃ Character showcase content built on existing performance footage The character sheet answers who's dancing. The depth map answers how - and keeping those two questions separate is the whole trick.

THE DEPTH MAP TRICK THAT FIXED DANCE ACCURACY IN SEEDANCE 2.0 Feed the model a video of someone dancing and it tries to interpret everything- the person, the clothes, the lighting, the room, and somewhere in there, the movement. Feed it a depth map and there's nothing left to interpret but the motion. Most creators trying to transfer a dance to a character reference the source footage directly, then wonder why the choreography drifts. The problem isn't the model - it's that you handed it ten variables when you only wanted one. Here's the workflow 1. Lock the character reference in GPT Image 2 first -face, build, costume, so identity holds independently of whatever motion gets applied to it 2. Convert the source dance footage into a depth map instead of using the raw video -this strips out the original performer's appearance, clothing, and environment entirely 3. Feed the depth map as the motion reference and the character sheet as the identity reference- two separate inputs doing two separate jobs, not one input trying to do both 5. Let the depth map carry only spatial movement -the model receives body position and momentum with no competing information about who's moving or what they look like 6. Keep the character and motion inputs isolated throughout - the moment you mix appearance data into the motion reference, the model starts negotiating between two identities Why this works • Raw footage passes the model everything at once- performer, wardrobe, room, lighting -and the choreography competes with all of it for attention • A depth map is pure spatial information, so the only thing left to transfer is movement • Separating identity from motion means the character can stay locked while the dance stays accurate - normally you're trading one for the other • The accuracy gain isn't the model getting better, it's the model getting fewer decisions to make Use cases: ⁃ Dance and choreography transfer onto original characters ⁃ Motion capture-style workflows without motion capture ⁃ Any sequence where a specific movement needs to survive intact ⁃ Character showcase content built on existing performance footage The character sheet answers who's dancing. The depth map answers how - and keeping those two questions separate is the whole trick.

Nexlow

116,680 次观看 • 9 天前

THE DEPTH MAP TRICK THAT FIXED DANCE ACCURACY IN SEEDANCE 2.0 Feed the model a video of someone dancing and it tries to interpret everything- the person, the clothes, the lighting, the room, and somewhere in there, the movement. Feed it a depth map and there's nothing left to interpret but the motion. Most creators trying to transfer a dance to a character reference the source footage directly, then wonder why the choreography drifts. The problem isn't the model - it's that you handed it ten variables when you only wanted one. Here's the workflow 1. Lock the character reference in GPT Image 2 first -face, build, costume, so identity holds independently of whatever motion gets applied to it 2. Convert the source dance footage into a depth map instead of using the raw video -this strips out the original performer's appearance, clothing, and environment entirely 3. Feed the depth map as the motion reference and the character sheet as the identity reference- two separate inputs doing two separate jobs, not one input trying to do both 5. Let the depth map carry only spatial movement -the model receives body position and momentum with no competing information about who's moving or what they look like 6. Keep the character and motion inputs isolated throughout - the moment you mix appearance data into the motion reference, the model starts negotiating between two identities Why this works • Raw footage passes the model everything at once- performer, wardrobe, room, lighting -and the choreography competes with all of it for attention • A depth map is pure spatial information, so the only thing left to transfer is movement • Separating identity from motion means the character can stay locked while the dance stays accurate - normally you're trading one for the other • The accuracy gain isn't the model getting better, it's the model getting fewer decisions to make Use cases: ⁃ Dance and choreography transfer onto original characters ⁃ Motion capture-style workflows without motion capture ⁃ Any sequence where a specific movement needs to survive intact ⁃ Character showcase content built on existing performance footage The character sheet answers who's dancing. The depth map answers how - and keeping those two questions separate is the whole trick.

THE DEPTH MAP TRICK THAT FIXED DANCE ACCURACY IN SEEDANCE 2.0 Feed the model a video of someone dancing and it tries to interpret everything- the person, the clothes, the lighting, the room, and somewhere in there, the movement. Feed it a depth map and there's nothing left to interpret but the motion. Most creators trying to transfer a dance to a character reference the source footage directly, then wonder why the choreography drifts. The problem isn't the model - it's that you handed it ten variables when you only wanted one. Here's the workflow 1. Lock the character reference in GPT Image 2 first -face, build, costume, so identity holds independently of whatever motion gets applied to it 2. Convert the source dance footage into a depth map instead of using the raw video -this strips out the original performer's appearance, clothing, and environment entirely 3. Feed the depth map as the motion reference and the character sheet as the identity reference- two separate inputs doing two separate jobs, not one input trying to do both 5. Let the depth map carry only spatial movement -the model receives body position and momentum with no competing information about who's moving or what they look like 6. Keep the character and motion inputs isolated throughout - the moment you mix appearance data into the motion reference, the model starts negotiating between two identities Why this works • Raw footage passes the model everything at once- performer, wardrobe, room, lighting -and the choreography competes with all of it for attention • A depth map is pure spatial information, so the only thing left to transfer is movement • Separating identity from motion means the character can stay locked while the dance stays accurate - normally you're trading one for the other • The accuracy gain isn't the model getting better, it's the model getting fewer decisions to make Use cases: ⁃ Dance and choreography transfer onto original characters ⁃ Motion capture-style workflows without motion capture ⁃ Any sequence where a specific movement needs to survive intact ⁃ Character showcase content built on existing performance footage The character sheet answers who's dancing. The depth map answers how - and keeping those two questions separate is the whole trick.

Nexlow

85,768 次观看 • 1 个月前

another men falling in love with carlos, color me shocked carlos: before we(the sport) didn't have a face, like it was just the car with the driver freddie: oh! you have a face carlos.. just looking at his eyes, he's beautiful isn't him 😭😭

another men falling in love with carlos, color me shocked carlos: before we(the sport) didn't have a face, like it was just the car with the driver freddie: oh! you have a face carlos.. just looking at his eyes, he's beautiful isn't him 😭😭

ari

160,158 次观看 • 1 年前

Jaycinco thought Lala Baptiste was OVERLY reacting by gasping to a scene in Euphoria showing her WORST fear, so he recorded her while having the most neutral face 😭🥀

Jaycinco thought Lala Baptiste was OVERLY reacting by gasping to a scene in Euphoria showing her WORST fear, so he recorded her while having the most neutral face 😭🥀

CutnPaid

95,751 次观看 • 17 天前

even though it's also not very good this series works SO MUCH better than rent a girlfriend does BC THE MC IS NOT A COMPLETE LOSER WITH NO REDEEMING QUALITIES 😭

even though it's also not very good this series works SO MUCH better than rent a girlfriend does BC THE MC IS NOT A COMPLETE LOSER WITH NO REDEEMING QUALITIES 😭

𝐌𝐢𝐥𝐤 ✿

29,381 次观看 • 1 年前

If you start punching cows in The Blood of Dawnwalker, they all die instantly. 💀 Coen isn't even a vampire yet, how is he so strong 😭

If you start punching cows in The Blood of Dawnwalker, they all die instantly. 💀 Coen isn't even a vampire yet, how is he so strong 😭

𝕯𝖗𝖊𝖆𝖉𝖗𝖔𝖇𝖊𝖗𝖙𝖘

43,607 次观看 • 7 天前

it's so embarrassing man 😭😭 someone please let him click a picture with Messi bro is trying hard.

it's so embarrassing man 😭😭 someone please let him click a picture with Messi bro is trying hard.

Gagan Choudhary

43,994 次观看 • 8 个月前

Diego is legit better than all our wingers bar Estevao and even more physical. His dribbling isn't even his best quality. It's his crossing. He s so accurate with them like one of the best wingers in the world with them. On the money Everytime.

Diego is legit better than all our wingers bar Estevao and even more physical. His dribbling isn't even his best quality. It's his crossing. He s so accurate with them like one of the best wingers in the world with them. On the money Everytime.

Oluwaseun 🇳🇬

38,590 次观看 • 7 个月前

$PI isn't just a coin, it's the future of global shopping! Everyone can Search, Buy & Sell everything with Map of Pi, paying with Pi directly between Pi Wallet in the Pi Browser. Have you shared #MapofPi🗺️ with your friends? Visit our app online

$PI isn't just a coin, it's the future of global shopping! Everyone can Search, Buy & Sell everything with Map of Pi, paying with Pi directly between Pi Wallet in the Pi Browser. Have you shared #MapofPi🗺️ with your friends? Visit our app online

Map Of Pi 🗺️

27,476 次观看 • 1 年前

We all want yeonbin pepero game but let’s be real, yeonjun won’t even last for a second😭 This is him when face to face with soobin😂

Sensitive content

We all want yeonbin pepero game but let’s be real, yeonjun won’t even last for a second😭 This is him when face to face with soobin😂

pi🏡❤️‍🩹

38,851 次观看 • 1 年前