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。

HOT! SCAIL-2 just dropped! End-to-end character animation via in-context conditioning - no skeleton middleman, it copies pixels directly - no glitches, no messy hands - 512p/704p - unified architecture for character replacement and multi-character tasks. - zero-shot generalization to animal-driven and mesh-based control.

Wildminder

10,953 subscribers

46,070 次观看 • 2 个月前 •via X (Twitter)

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

Asmongold gives his biggest advice to video game studios thinking about making female characters following the Xbox Studios layoff situation 👀 “If you're ever thinking about making a female character in a video game, ask yourself, would girls that have onlyfans cosplay this character? If the answer to that question is no, delete the character. That's it. That's the beginning and end of it”

Asmongold gives his biggest advice to video game studios thinking about making female characters following the Xbox Studios layoff situation 👀 “If you're ever thinking about making a female character in a video game, ask yourself, would girls that have onlyfans cosplay this character? If the answer to that question is no, delete the character. That's it. That's the beginning and end of it”

Prism

262,432 次观看 • 1 个月前

Meet Grizelda! (4K link below) A short animation I've been working on for quite some time. I hired a group of artists to create this character! From concept to modeling and rigging it! Credits Creator & Animator: Jonathan Veiga Voice Actor: OolayTiger Character Designer: Zeker_Art Character Model: Bahman Zarei Grooming Artist: Christina De Luca Character Rigger: alargewoodenbadger Music: No More Heroes 2 OST: Sound's Asylum The background was all built manually, but I bought assets in the marketplace in Unreal to help me build the world faster. I also did all the lighting!

Meet Grizelda! (4K link below) A short animation I've been working on for quite some time. I hired a group of artists to create this character! From concept to modeling and rigging it! Credits Creator & Animator: Jonathan Veiga Voice Actor: OolayTiger Character Designer: Zeker_Art Character Model: Bahman Zarei Grooming Artist: Christina De Luca Character Rigger: alargewoodenbadger Music: No More Heroes 2 OST: Sound's Asylum The background was all built manually, but I bought assets in the marketplace in Unreal to help me build the world faster. I also did all the lighting!

Jonathan Veiga

270,880 次观看 • 2 年前

✦ Path to Inner Light | Margie · Cityscape Roamer "What? You've already heard of my title 'Princess Rubbish Bin'?! Ugh... please don't go along with it! Just call me Margie. No princess here, just a little scavenger of Icelake!" Margie's Character Animation Showcase VA: Carina Reeves ✦ How to Obtain Characters Free via Covert Commissions / Purchase Character Shards ✦ Follow the DuetNightAbyss & Retweet this post and we will select 5 winners to receive a Random Character Acrylic Standee. Event Ends: December 15 ✦ Download Duet Night Abyss #DuetNightAbyss #Margie

✦ Path to Inner Light | Margie · Cityscape Roamer "What? You've already heard of my title 'Princess Rubbish Bin'?! Ugh... please don't go along with it! Just call me Margie. No princess here, just a little scavenger of Icelake!" Margie's Character Animation Showcase VA: Carina Reeves ✦ How to Obtain Characters Free via Covert Commissions / Purchase Character Shards ✦ Follow the DuetNightAbyss & Retweet this post and we will select 5 winners to receive a Random Character Acrylic Standee. Event Ends: December 15 ✦ Download Duet Night Abyss #DuetNightAbyss #Margie

DuetNightAbyss

14,504 次观看 • 8 个月前

AI animation isn't hard. You're probably just making it harder than it needs to be. I made this animated short in under an hour using SeaArt.Ai🐋 and a simple workflow I call: "The Cut Method" No complicated workflows. No dozens of tools. No advanced editing. Just character sheets, one environment, and a series of connected Seedance clips. Workflow below 🧵⬇️

AI animation isn't hard. You're probably just making it harder than it needs to be. I made this animated short in under an hour using SeaArt.Ai🐋 and a simple workflow I call: "The Cut Method" No complicated workflows. No dozens of tools. No advanced editing. Just character sheets, one environment, and a series of connected Seedance clips. Workflow below 🧵⬇️

jey jey ⚜️

24,818 次观看 • 2 个月前

Everyone on X has been buzzing about character consistency in 2024. We’re excited to present: multi-character consistency! With a simple workflow, you can easily train custom models that generate multiple consistent characters (2 or more) right from your browser on #Scenario—no complex setup or editing needed. They can take on authentic poses and actions based on a simple prompt. Check out the short video below and more details!

Everyone on X has been buzzing about character consistency in 2024. We’re excited to present: multi-character consistency! With a simple workflow, you can easily train custom models that generate multiple consistent characters (2 or more) right from your browser on #Scenario—no complex setup or editing needed. They can take on authentic poses and actions based on a simple prompt. Check out the short video below and more details!

Emm | scenario.com

21,410 次观看 • 1 年前

No, he without a doubt, died in Berlin. He considered anyone fleeing to be a coward. He was a fanatic, it would have been completely of character for him to flee and give up the country he loved and fought for just to live in Argentina.

No, he without a doubt, died in Berlin. He considered anyone fleeing to be a coward. He was a fanatic, it would have been completely of character for him to flee and give up the country he loved and fought for just to live in Argentina.

Mason

2,754,735 次观看 • 2 个月前

$85k+/month creatives are coming from animated supplements like this a lymphatic drainage bottle turned into a character arms flexing face smiling energy radiating it’s not a static product shot it’s a personality the visual exaggerates the benefit the character makes the claim memorable the animation holds attention longer people don’t scroll past it they watch to see what it does next no influencer needed no before and after photos no complicated production just one animated object running inside a structured ad system that’s why these convert so well rt + comment “lymph” and i’ll send the framework (follow for dm)

$85k+/month creatives are coming from animated supplements like this a lymphatic drainage bottle turned into a character arms flexing face smiling energy radiating it’s not a static product shot it’s a personality the visual exaggerates the benefit the character makes the claim memorable the animation holds attention longer people don’t scroll past it they watch to see what it does next no influencer needed no before and after photos no complicated production just one animated object running inside a structured ad system that’s why these convert so well rt + comment “lymph” and i’ll send the framework (follow for dm)

Florin

30,524 次观看 • 5 个月前

CHINESE GIRL JUST DROPPED A 52-SECOND AI VIDEO GUIDE No camera crew. No reshoots. No avatar studio. She opens Tongyi, picks digital person -> character replacement, uploads one source clip + one reference photo, hits generate. The new video keeps the same expressions, eye movement and gestures. This is how people are going to mass-produce AI creator content now. Save this before somebody turns it into a $99 course.

0xTria

75,309 次观看 • 1 个月前

I found a tool that turns a single character image into fully animated sprite sheets in seconds. AI just killed the most annoying part of game development. And honestly this is getting kind of crazy 👇 AutoSprite uses AI to generate walk cycles, attacks, idles, and basically whatever moveset you need. What's new: → Character pose editor → Cutscene creator → Best-in-class background remover → Higher resolutions → Faster generations You upload one image. It does the rest. No frame-by-frame drawing. No animation pipeline headache. No animation experience needed. Solo devs, studios, and anyone prototyping fast, this cuts hours of work down to seconds. Exports drop straight into Unity, Godot, GameMaker, Phaser, and RPG Maker. Zero extra setup. Try it free at

I found a tool that turns a single character image into fully animated sprite sheets in seconds. AI just killed the most annoying part of game development. And honestly this is getting kind of crazy 👇 AutoSprite uses AI to generate walk cycles, attacks, idles, and basically whatever moveset you need. What's new: → Character pose editor → Cutscene creator → Best-in-class background remover → Higher resolutions → Faster generations You upload one image. It does the rest. No frame-by-frame drawing. No animation pipeline headache. No animation experience needed. Solo devs, studios, and anyone prototyping fast, this cuts hours of work down to seconds. Exports drop straight into Unity, Godot, GameMaker, Phaser, and RPG Maker. Zero extra setup. Try it free at

Jafar Najafov

35,410 次观看 • 1 个月前

NVIDIA just dropped another revolution in AI animation — real-time and open source. It's called ARDY. You type what the character should do — and it just does it. While moving. Change the text mid-walk, click a waypoint, add keyframes — it re-plans instantly. No baking, no waiting. And this time they dropped EVERYTHING at once: code, weights, working demo. Fully ready to run locally. I tested it. Honestly — shock. Feels like Kimodo × MotionBricks: that amount of control (text + constraints + trajectories), but live. And it fits in 20–24GB of VRAM. I didn't expect animation to move this fast — I thought 3D would get solved first. It's robotics pushing it: the same model drives their humanoid robots. Paper + code:

NVIDIA just dropped another revolution in AI animation — real-time and open source. It's called ARDY. You type what the character should do — and it just does it. While moving. Change the text mid-walk, click a waypoint, add keyframes — it re-plans instantly. No baking, no waiting. And this time they dropped EVERYTHING at once: code, weights, working demo. Fully ready to run locally. I tested it. Honestly — shock. Feels like Kimodo × MotionBricks: that amount of control (text + constraints + trajectories), but live. And it fits in 20–24GB of VRAM. I didn't expect animation to move this fast — I thought 3D would get solved first. It's robotics pushing it: the same model drives their humanoid robots. Paper + code:

Stefan 3D AI

110,180 次观看 • 1 个月前

I couldn't believe it. One product image = a full video ad with Wizstar. No script. No shoot. No editor. Just the image, one sentence of instruction, and Wizstar did everything else. It analyzes your product automatically, builds the scene around it, and generates a complete marketing video. Character stays consistent across every shot. Product details stay accurate. Audio and video sync perfectly. It powers multi-model invocation to match any style you need, and Seedance 2.0 supports face input for consistent character generation. But wait, it gets better. You can also upload any viral video from TikTok or Amazon. Then it will recreate that EXACT winning formula for YOUR product. No more guessing what works. Zero creative trial and error. Make a video today, start seeing sales in 7 days. New users get free credits on signup. First subscription is only $19 and includes a 30-second E-commerce Agent experience. Try it here: Who's making their first ad tonight? 👇 #Wizstar #AIVideo #EcommerceMarketing

I couldn't believe it. One product image = a full video ad with Wizstar. No script. No shoot. No editor. Just the image, one sentence of instruction, and Wizstar did everything else. It analyzes your product automatically, builds the scene around it, and generates a complete marketing video. Character stays consistent across every shot. Product details stay accurate. Audio and video sync perfectly. It powers multi-model invocation to match any style you need, and Seedance 2.0 supports face input for consistent character generation. But wait, it gets better. You can also upload any viral video from TikTok or Amazon. Then it will recreate that EXACT winning formula for YOUR product. No more guessing what works. Zero creative trial and error. Make a video today, start seeing sales in 7 days. New users get free credits on signup. First subscription is only $19 and includes a 30-second E-commerce Agent experience. Try it here: Who's making their first ad tonight? 👇 #Wizstar #AIVideo #EcommerceMarketing

Pradeep Pandey

59,917 次观看 • 2 个月前

Growing up, there was no Sleepaway camp in my house. My father just tossed me in the backyard for three months and told me to “look for worms” said it built character. Honestly… he wasn’t wrong. My new special ‘It Ain’t Right’ is airing on Hulu and Disney+ this Friday

Growing up, there was no Sleepaway camp in my house. My father just tossed me in the backyard for three months and told me to “look for worms” said it built character. Honestly… he wasn’t wrong. My new special ‘It Ain’t Right’ is airing on Hulu and Disney+ this Friday

Sebastian Maniscalco

53,822 次观看 • 8 个月前

You can now generate 25-second Sora 2 videos in a single shot 🤯 No clipping, no stitching. One full, 25-second generation. Most Sora 2 providers cap you at 10-15 seconds. This model lets you go up to 25, with scene-by-scene control. Perfect for e-comm brands & agencies creating AI UGC that doesn't require a dozen edits. Here's the problem: You're generating 10-second Sora 2 clips and stitching them together in post. And it takes FOREVER. This Sora 2 Pro Storyboard model fixes it: → Specify up to 6 different scenes → Set the exact duration for each scene → Upload a product reference image → AI generates all scenes in one continuous output → 25 seconds of cohesive UGC in a single shot No choppy edits. No character inconsistency between clips. No post-production stitching. What you get: - Full 25-second AI UGC videos - Scene-by-scene control over the narrative - Consistent character and setting throughout - $1.35 per video I recorded a 10-minute walkthrough showing exactly how to set this up step-by-step. Want access? > Comment "25" > Like this post And I'll send it over (must be following so I can DM)

You can now generate 25-second Sora 2 videos in a single shot 🤯 No clipping, no stitching. One full, 25-second generation. Most Sora 2 providers cap you at 10-15 seconds. This model lets you go up to 25, with scene-by-scene control. Perfect for e-comm brands & agencies creating AI UGC that doesn't require a dozen edits. Here's the problem: You're generating 10-second Sora 2 clips and stitching them together in post. And it takes FOREVER. This Sora 2 Pro Storyboard model fixes it: → Specify up to 6 different scenes → Set the exact duration for each scene → Upload a product reference image → AI generates all scenes in one continuous output → 25 seconds of cohesive UGC in a single shot No choppy edits. No character inconsistency between clips. No post-production stitching. What you get: - Full 25-second AI UGC videos - Scene-by-scene control over the narrative - Consistent character and setting throughout - $1.35 per video I recorded a 10-minute walkthrough showing exactly how to set this up step-by-step. Want access? > Comment "25" > Like this post And I'll send it over (must be following so I can DM)

Mike Futia

15,329 次观看 • 7 个月前

Alibaba presents MIMO Controllable Character Video Synthesis with Spatial Decomposed Modeling Character video synthesis aims to produce realistic videos of animatable characters within lifelike scenes. As a fundamental problem in the computer vision and graphics community, 3D works typically require multi-view captures for per-case training, which severely limits their applicability of modeling arbitrary characters in a short time. Recent 2D methods break this limitation via pre-trained diffusion models, but they struggle for pose generality and scene interaction. To this end, we propose MIMO, a novel framework which can not only synthesize character videos with controllable attributes (i.e., character, motion and scene) provided by simple user inputs, but also simultaneously achieve advanced scalability to arbitrary characters, generality to novel 3D motions, and applicability to interactive real-world scenes in a unified framework. The core idea is to encode the 2D video to compact spatial codes, considering the inherent 3D nature of video occurrence. Concretely, we lift the 2D frame pixels into 3D using monocular depth estimators, and decompose the video clip to three spatial components (i.e., main human, underlying scene, and floating occlusion) in hierarchical layers based on the 3D depth. These components are further encoded to canonical identity code, structured motion code and full scene code, which are utilized as control signals of synthesis process. The design of spatial decomposed modeling enables flexible user control, complex motion expression, as well as 3D-aware synthesis for scene interactions. Experimental results demonstrate effectiveness and robustness of the proposed method.

Alibaba presents MIMO Controllable Character Video Synthesis with Spatial Decomposed Modeling Character video synthesis aims to produce realistic videos of animatable characters within lifelike scenes. As a fundamental problem in the computer vision and graphics community, 3D works typically require multi-view captures for per-case training, which severely limits their applicability of modeling arbitrary characters in a short time. Recent 2D methods break this limitation via pre-trained diffusion models, but they struggle for pose generality and scene interaction. To this end, we propose MIMO, a novel framework which can not only synthesize character videos with controllable attributes (i.e., character, motion and scene) provided by simple user inputs, but also simultaneously achieve advanced scalability to arbitrary characters, generality to novel 3D motions, and applicability to interactive real-world scenes in a unified framework. The core idea is to encode the 2D video to compact spatial codes, considering the inherent 3D nature of video occurrence. Concretely, we lift the 2D frame pixels into 3D using monocular depth estimators, and decompose the video clip to three spatial components (i.e., main human, underlying scene, and floating occlusion) in hierarchical layers based on the 3D depth. These components are further encoded to canonical identity code, structured motion code and full scene code, which are utilized as control signals of synthesis process. The design of spatial decomposed modeling enables flexible user control, complex motion expression, as well as 3D-aware synthesis for scene interactions. Experimental results demonstrate effectiveness and robustness of the proposed method.

AK

149,079 次观看 • 1 年前

Seedance 2.5 is coming to If you've ever stitched 5 clips together praying they'd match — this is for you. Native 30s in a single pass in crisp 4K, with audio generated directly alongside the video — not slapped on after. Up to 50 references to ensure your character and style hold across the entire shot. One take. No duct tape. 🙌

Seedance 2.5 is coming to If you've ever stitched 5 clips together praying they'd match — this is for you. Native 30s in a single pass in crisp 4K, with audio generated directly alongside the video — not slapped on after. Up to 50 references to ensure your character and style hold across the entire shot. One take. No duct tape. 🙌

Medeo_AI

549,901 次观看 • 14 天前

MotionBricks by NVIDIA. the all-in-one brain for character movement. No manual rigging/animation, you just let this AI handle the physics and style in real-time. - 15k FPS on RTX 5090! You could run an entire city of unique NPCs without breaking a sweat. - Smart Primitives. The AI fills in the natural-looking motion automatically. - Zero-Shot Skills. - Glitch-Free. Keeps movements stable.

MotionBricks by NVIDIA. the all-in-one brain for character movement. No manual rigging/animation, you just let this AI handle the physics and style in real-time. - 15k FPS on RTX 5090! You could run an entire city of unique NPCs without breaking a sweat. - Smart Primitives. The AI fills in the natural-looking motion automatically. - Zero-Shot Skills. - Glitch-Free. Keeps movements stable.

Wildminder

23,296 次观看 • 1 个月前

Prompting for Seedance 2.0 tip 🫆 It doesn't have to be complex to give good results. You can get surprisingly far with a good reference image + a simple narrative prompt. In this video, i combined 2 Seedance 2.0 generations (prompts below): Part 1: "an animator is struggling to come up with ideas. it's driving him made. he throws out the drawing. tries again. pulls his beard. starts over. paper thrown into trash can. comedy. animation style. multi-scene. high end animation." Part 2 (giving it part 1 video as reference) "continue this video. after struggling to come up with a concept for a character, the artist take a deep breath and say to him self to focus harold. then he draws a beautiful and cool badass princess with a gun. he holds it up proud and whispers that's it. comedy. animation style. multi-scene. high end animation."

Prompting for Seedance 2.0 tip 🫆 It doesn't have to be complex to give good results. You can get surprisingly far with a good reference image + a simple narrative prompt. In this video, i combined 2 Seedance 2.0 generations (prompts below): Part 1: "an animator is struggling to come up with ideas. it's driving him made. he throws out the drawing. tries again. pulls his beard. starts over. paper thrown into trash can. comedy. animation style. multi-scene. high end animation." Part 2 (giving it part 1 video as reference) "continue this video. after struggling to come up with a concept for a character, the artist take a deep breath and say to him self to focus harold. then he draws a beautiful and cool badass princess with a gun. he holds it up proud and whispers that's it. comedy. animation style. multi-scene. high end animation."

Martin LeBlanc

30,942 次观看 • 1 个月前

We just released our new Character Control Rig Addon for Blender! We listened to our user's requests, and this was created specifically to help them with their MetaHuman animation workflows in Blender. Want to learn more? We just dropped a new tutorial demoing how you can use it with Rigify, Mixamo, and Unreal's robust Retargetor. Importing and exporting animations onto FK controls and the MetaHuman Face Board is dead simple now. #BlenderTutorial #MetaHuman #UE5 #blender

We just released our new Character Control Rig Addon for Blender! We listened to our user's requests, and this was created specifically to help them with their MetaHuman animation workflows in Blender. Want to learn more? We just dropped a new tutorial demoing how you can use it with Rigify, Mixamo, and Unreal's robust Retargetor. Importing and exporting animations onto FK controls and the MetaHuman Face Board is dead simple now. #BlenderTutorial #MetaHuman #UE5 #blender

Poly Hammer

11,742 次观看 • 3 个月前

Phantom Blade Zero Info: -NO soulslike game -Combo-driven -There will be difficulty options -Rich character progression -No respawn for mobs after checkpoints or death -Unreal Engine 5 -2 graphic modes on PS5 -Fluid 60fps target -Secret bosses -Making the most out of the DualSense -Each weapon affects gameplay -Based on Chinese Wuxia culture -Range of cool and dark elements -New style called "KungfuPunk" -Inspired by Ninja Gaiden, Sekiro and Nioh -Multiplayer endgame content planned -Coming to PS5 and PC -Independent game studio in China (S-GAME) -No release date yet -Release date coming later this year

Phantom Blade Zero Info: -NO soulslike game -Combo-driven -There will be difficulty options -Rich character progression -No respawn for mobs after checkpoints or death -Unreal Engine 5 -2 graphic modes on PS5 -Fluid 60fps target -Secret bosses -Making the most out of the DualSense -Each weapon affects gameplay -Based on Chinese Wuxia culture -Range of cool and dark elements -New style called "KungfuPunk" -Inspired by Ninja Gaiden, Sekiro and Nioh -Multiplayer endgame content planned -Coming to PS5 and PC -Independent game studio in China (S-GAME) -No release date yet -Release date coming later this year

GermanStrands

921,563 次观看 • 1 年前

The final lecture of my course is an intro to character training! This is a topic that I've been quietly very invested in for ~18 months, as it: * Has potential for high real world impact * Clearly used extensively at frontier labs * Almost no empirical literature exists * More accessible on academic compute This lecture covers what character training is, reviews model specs, constitutions, the differences, the motivations in real world events, some example research papers I like, and open questions in how it relates to post-training/model use generally. Hopefully this brings more people into the field (and reach out if you have questions). It is one of the more research-y chapters in my book, but one that I felt needed the reference. There is still so little, educational content on the topic online. 0:00 Intro 6:22 Part 1: Fundamentals — character, constitutions, and model specs 19:21 Part 2: Character training in practice 23:23 Part 3: Character elicitation without gradient steps 28:03 Part 4: Open questions (and the end of the course) 32:27 The course, complete Thanks for watching. No need to like and subscribe now that the course is done, you definitely wouldn't! h/t to Sharan for leading the technical work I got to do in the space, and Zafir Stojanovski for investing a lot of attention at this book chapter.

The final lecture of my course is an intro to character training! This is a topic that I've been quietly very invested in for ~18 months, as it: * Has potential for high real world impact * Clearly used extensively at frontier labs * Almost no empirical literature exists * More accessible on academic compute This lecture covers what character training is, reviews model specs, constitutions, the differences, the motivations in real world events, some example research papers I like, and open questions in how it relates to post-training/model use generally. Hopefully this brings more people into the field (and reach out if you have questions). It is one of the more research-y chapters in my book, but one that I felt needed the reference. There is still so little, educational content on the topic online. 0:00 Intro 6:22 Part 1: Fundamentals — character, constitutions, and model specs 19:21 Part 2: Character training in practice 23:23 Part 3: Character elicitation without gradient steps 28:03 Part 4: Open questions (and the end of the course) 32:27 The course, complete Thanks for watching. No need to like and subscribe now that the course is done, you definitely wouldn't! h/t to Sharan for leading the technical work I got to do in the space, and Zafir Stojanovski for investing a lot of attention at this book chapter.

Nathan Lambert

56,080 次观看 • 7 天前