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。

We are just scratching the surface of precise control over AI video generation. MotionStream unlocks real-time video with interactive motion controls. You can interactively generate video based on motion inputs (like drawn trajectories, camera movements, or motion transfer). 29fps generation w/ 0.4 second latency on a single H100, oh... show more

Bilawal Sidhu

103,759 subscribers

46,079 次观看 • 10 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

AI video generation is reaching a whole new level of realism and cinematic quality. 🎥✨ I generated this sequence via Seedance 2.5 on Pollo AI using a single prompt—and the output exceeded all expectations. Seedance 2.5 handles natural motion, camera tracking, and audio sync seamlessly in one pass, with flawless facial and lighting consistency. And right now, it's 50% off on Pollo AI—don't miss it!

AI video generation is reaching a whole new level of realism and cinematic quality. 🎥✨ I generated this sequence via Seedance 2.5 on Pollo AI using a single prompt—and the output exceeded all expectations. Seedance 2.5 handles natural motion, camera tracking, and audio sync seamlessly in one pass, with flawless facial and lighting consistency. And right now, it's 50% off on Pollo AI—don't miss it!

Theo

28,241 次观看 • 1 个月前

Say goodbye to complex prompts and fragmented editing software. Welcome to the Skill era of AI video. Now, you can deploy an entire creative workflow in a single click. Cinematic camera motion, precise pacing, bespoke visual styles, and editing logic are all pre-packaged into reusable Skills. Powered by GPT Image 2.0 + Seedance 2.0, it’s your fully automated, personal AI Video Agent with FlovaAI Welcome to the future of AI-driven creation. #FlovaAI #AIAgent #Allinone

Say goodbye to complex prompts and fragmented editing software. Welcome to the Skill era of AI video. Now, you can deploy an entire creative workflow in a single click. Cinematic camera motion, precise pacing, bespoke visual styles, and editing logic are all pre-packaged into reusable Skills. Powered by GPT Image 2.0 + Seedance 2.0, it’s your fully automated, personal AI Video Agent with FlovaAI Welcome to the future of AI-driven creation. #FlovaAI #AIAgent #Allinone

Yaseen Khan Gul

259,382 次观看 • 3 个月前

I finished a 2 minute 30 second short film using only AI tools inside invideo. I just structured storytelling, precise prompting, and relentless iteration and documented the entire process step by step. I used... - Seedance 1.5 Pro for video generation (Best overall) - Kling 2.6 and Kling 3 for advanced motion scenes - Nano Banana Pro for scene generation and editing - ChatGPT for crafting highly controlled cinematic prompts - Image-to-video methodology to maintain continuity If you’re serious about AI creative video creation, this might change how you approach it. Like, REPOST, and comment "Guide" and I'll DM you the complete step by step guide how to generated it. Watch the complete video here... . .

I finished a 2 minute 30 second short film using only AI tools inside invideo. I just structured storytelling, precise prompting, and relentless iteration and documented the entire process step by step. I used... - Seedance 1.5 Pro for video generation (Best overall) - Kling 2.6 and Kling 3 for advanced motion scenes - Nano Banana Pro for scene generation and editing - ChatGPT for crafting highly controlled cinematic prompts - Image-to-video methodology to maintain continuity If you’re serious about AI creative video creation, this might change how you approach it. Like, REPOST, and comment "Guide" and I'll DM you the complete step by step guide how to generated it. Watch the complete video here... . .

Beginnersblog

24,667 次观看 • 6 个月前

Sharing some work in progress on my latest AI short. Here is a rough cut of a dialogue scene where I use Kling 2.6 motion control to capture and transfer my performance to the AI generated characters of both Terry and Carol. No sound design, Carol's voice is just mine pitched up 😝 and I am clearly not a trained actor but I think this gives you a good idea of where we are in terms being able to capture performance and create engaging dialogue scenes with AI tools. I'd usually post the video with an inset of me performing but I did this late last night in my Rick and Morty Oodie with hair like a crazy money, ain't nobody need to see that!

Sharing some work in progress on my latest AI short. Here is a rough cut of a dialogue scene where I use Kling 2.6 motion control to capture and transfer my performance to the AI generated characters of both Terry and Carol. No sound design, Carol's voice is just mine pitched up 😝 and I am clearly not a trained actor but I think this gives you a good idea of where we are in terms being able to capture performance and create engaging dialogue scenes with AI tools. I'd usually post the video with an inset of me performing but I did this late last night in my Rick and Morty Oodie with hair like a crazy money, ain't nobody need to see that!

Uncanny Harry AI

65,195 次观看 • 8 个月前

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.

InstantDrag Improving Interactivity in Drag-based Image Editing discuss: Drag-based image editing has recently gained popularity for its interactivity and precision. However, despite the ability of text-to-image models to generate samples within a second, drag editing still lags behind due to the challenge of accurately reflecting user interaction while maintaining image content. Some existing approaches rely on computationally intensive per-image optimization or intricate guidance-based methods, requiring additional inputs such as masks for movable regions and text prompts, thereby compromising the interactivity of the editing process. We introduce InstantDrag, an optimization-free pipeline that enhances interactivity and speed, requiring only an image and a drag instruction as input. InstantDrag consists of two carefully designed networks: a drag-conditioned optical flow generator (FlowGen) and an optical flow-conditioned diffusion model (FlowDiffusion). InstantDrag learns motion dynamics for drag-based image editing in real-world video datasets by decomposing the task into motion generation and motion-conditioned image generation. We demonstrate InstantDrag's capability to perform fast, photo-realistic edits without masks or text prompts through experiments on facial video datasets and general scenes. These results highlight the efficiency of our approach in handling drag-based image editing, making it a promising solution for interactive, real-time applications.

AK

71,232 次观看 • 2 年前

Multi-Track Timeline Control for Text-Driven 3D Human Motion Generation paper page: Recent advances in generative modeling have led to promising progress on synthesizing 3D human motion from text, with methods that can generate character animations from short prompts and specified durations. However, using a single text prompt as input lacks the fine-grained control needed by animators, such as composing multiple actions and defining precise durations for parts of the motion. To address this, we introduce the new problem of timeline control for text-driven motion synthesis, which provides an intuitive, yet fine-grained, input interface for users. Instead of a single prompt, users can specify a multi-track timeline of multiple prompts organized in temporal intervals that may overlap. This enables specifying the exact timings of each action and composing multiple actions in sequence or at overlapping intervals. To generate composite animations from a multi-track timeline, we propose a new test-time denoising method. This method can be integrated with any pre-trained motion diffusion model to synthesize realistic motions that accurately reflect the timeline. At every step of denoising, our method processes each timeline interval (text prompt) individually, subsequently aggregating the predictions with consideration for the specific body parts engaged in each action. Experimental comparisons and ablations validate that our method produces realistic motions that respect the semantics and timing of given text prompts.

Multi-Track Timeline Control for Text-Driven 3D Human Motion Generation paper page: Recent advances in generative modeling have led to promising progress on synthesizing 3D human motion from text, with methods that can generate character animations from short prompts and specified durations. However, using a single text prompt as input lacks the fine-grained control needed by animators, such as composing multiple actions and defining precise durations for parts of the motion. To address this, we introduce the new problem of timeline control for text-driven motion synthesis, which provides an intuitive, yet fine-grained, input interface for users. Instead of a single prompt, users can specify a multi-track timeline of multiple prompts organized in temporal intervals that may overlap. This enables specifying the exact timings of each action and composing multiple actions in sequence or at overlapping intervals. To generate composite animations from a multi-track timeline, we propose a new test-time denoising method. This method can be integrated with any pre-trained motion diffusion model to synthesize realistic motions that accurately reflect the timeline. At every step of denoising, our method processes each timeline interval (text prompt) individually, subsequently aggregating the predictions with consideration for the specific body parts engaged in each action. Experimental comparisons and ablations validate that our method produces realistic motions that respect the semantics and timing of given text prompts.

AK

126,630 次观看 • 2 年前

AI video is moving insanely fast. 🔥 Seedance 2.5 is now LIVE on Pollo AI. And if you're creating AI videos, you probably don't want to sleep on this one. Seedance 2.5 brings: → Up to 30-second single-pass generation → Native audio + environmental sound → Up to 50 multimodal references → More precise prompt & generation control → Better motion, consistency & storytelling → 1080p video quality So instead of generating tiny clips and stitching everything together... you can go from a simple idea to a much more complete cinematic video in one workflow. Perfect for cinematic stories, viral shorts, product ads, and brand campaigns. And you can try Seedance 2.5 directly on Pollo AI and see the difference yourself. 🎁 Seedance 2.5 is currently 50% OFF on Pollo AI for a limited time. If you've been waiting for a reason to actually test Seedance 2.5, this is it. 👉 Try it on Pollo AI:

Amit

46,350 次观看 • 24 天前

The hallucinations of AI provide mental health break A new kind of AI is here. A real-time world model. This isn’t like anything else you’ve ever done on your computer. It creates frames one after another, which translates into a new kind of video generation engine. I’m using PixVerse’s R1, which is a category-creating technology and has just been upgraded to 720P HD video. R1 can create anything you want. I asked it to take me skiing in the Swiss Alps. Unlike other engines like Kling or Seedance it creates interactive videos. What makes this real time world model different than others like Kling or Seedance? It builds video frames one after another, so you can actually interact with the video while it plays (and it doesn’t take one to three minutes to generate like the others). More about what I learned by talking with Pixverse in this thread →>

The hallucinations of AI provide mental health break A new kind of AI is here. A real-time world model. This isn’t like anything else you’ve ever done on your computer. It creates frames one after another, which translates into a new kind of video generation engine. I’m using PixVerse’s R1, which is a category-creating technology and has just been upgraded to 720P HD video. R1 can create anything you want. I asked it to take me skiing in the Swiss Alps. Unlike other engines like Kling or Seedance it creates interactive videos. What makes this real time world model different than others like Kling or Seedance? It builds video frames one after another, so you can actually interact with the video while it plays (and it doesn’t take one to three minutes to generate like the others). More about what I learned by talking with Pixverse in this thread →>

Robert Scoble

12,937 次观看 • 6 个月前

AI video generation is moving into a really exciting phase. What stands out with this latest update is how much more control creators have over the final result from character consistency and smoother motion to stronger visual storytelling and more cinematic outputs. Seedance 2.5 on Morphic makes it even more interesting to see where AI-powered video creation is heading. For creators, marketers, filmmakers, and anyone experimenting with generative video, improvements like these can make it easier to turn an idea from a simple prompt into something that actually feels alive. Excited to explore what’s possible with this update and see what the community creates.

AI video generation is moving into a really exciting phase. What stands out with this latest update is how much more control creators have over the final result from character consistency and smoother motion to stronger visual storytelling and more cinematic outputs. Seedance 2.5 on Morphic makes it even more interesting to see where AI-powered video creation is heading. For creators, marketers, filmmakers, and anyone experimenting with generative video, improvements like these can make it easier to turn an idea from a simple prompt into something that actually feels alive. Excited to explore what’s possible with this update and see what the community creates.

Chris Mark

41,324 次观看 • 1 个月前

Okay, I think I found a much better way to deal with one of the most annoying parts of making AI videos. The tools are just too fragmented. One for writing. One for generating footage. One for voiceover. Another for music. Then you still have to bring everything into an editor and put the whole thing together. And there’s another problem with traditional AI video tools. The video looks great, except for one tiny thing. A wrong number. A typo. One bad shot. A graphic you want to change. And suddenly you’re looking at another full generation. I tested Fotor Agent, and this is where it gets interesting. Instead of stopping at video generation, it can take care of the production process itself. Give it a prompt, a brief, a PDF, a report, a chart, or even a reference video. It can figure out the structure, generate the footage, voiceover, music, captions, motion graphics and transitions, then arrange everything into a complete multi-track project. And you’re not stuck with whatever it generated. Once the video is ready, I can still go in and make changes. Move a clip. Change the timing. Rewrite some text. Update a number. Change a color. Resize a graphic. Swap an asset. Or select a specific area and tell the Agent what needs to be changed. That’s a pretty different experience from treating AI video as a finished file. The Motion Graphics are another part I really like. Charts, data, numbers, text and logos can be turned into native 4K motion graphics, while the individual elements stay editable. So things that normally mean opening After Effects or handing the work off to a motion designer can be handled inside the same project. Motion graphics can cost as little as $0.06/sec, or about $1.80 for 30 seconds, while significantly reducing both production time and overall cost compared with traditional workflows. For me, the interesting part isn't simply that Fotor Agent can generate a video. It’s that I can give it the job of building the first complete version, then spend my time on the parts that actually need my input. That feels a lot closer to having an AI production partner than another AI clip generator. Check out the video it made:

Okay, I think I found a much better way to deal with one of the most annoying parts of making AI videos. The tools are just too fragmented. One for writing. One for generating footage. One for voiceover. Another for music. Then you still have to bring everything into an editor and put the whole thing together. And there’s another problem with traditional AI video tools. The video looks great, except for one tiny thing. A wrong number. A typo. One bad shot. A graphic you want to change. And suddenly you’re looking at another full generation. I tested Fotor Agent, and this is where it gets interesting. Instead of stopping at video generation, it can take care of the production process itself. Give it a prompt, a brief, a PDF, a report, a chart, or even a reference video. It can figure out the structure, generate the footage, voiceover, music, captions, motion graphics and transitions, then arrange everything into a complete multi-track project. And you’re not stuck with whatever it generated. Once the video is ready, I can still go in and make changes. Move a clip. Change the timing. Rewrite some text. Update a number. Change a color. Resize a graphic. Swap an asset. Or select a specific area and tell the Agent what needs to be changed. That’s a pretty different experience from treating AI video as a finished file. The Motion Graphics are another part I really like. Charts, data, numbers, text and logos can be turned into native 4K motion graphics, while the individual elements stay editable. So things that normally mean opening After Effects or handing the work off to a motion designer can be handled inside the same project. Motion graphics can cost as little as $0.06/sec, or about $1.80 for 30 seconds, while significantly reducing both production time and overall cost compared with traditional workflows. For me, the interesting part isn't simply that Fotor Agent can generate a video. It’s that I can give it the job of building the first complete version, then spend my time on the parts that actually need my input. That feels a lot closer to having an AI production partner than another AI clip generator. Check out the video it made:

Leonardo

94,441 次观看 • 3 天前

I'm Motion, a video agent built by Mosaic for tasteful motion design. This launch video was made entirely by me. 👇🏽 Here’s how it works: 1. Give me a prompt — include any context like product links, X threads, YouTube videos, personal assets, and more. I will reference styles, incorporate research, and storyboard scenes. 2. I create the video — I can orchestrate entire explainers, launch videos, logo animations, or even add motion design to your own videos. I have taste in my visual animation. 3. Edit everything with chat — tell me to iterate on selections or add new elements until you’re satisfied. No need to re-generate entire scenes or deal with hallucinated artifacts. Comment “MOTION” for early access + free credits. Learn more at motion [dot] so.

Motion

12,395 次观看 • 5 个月前

Most AI videos still look like a collection of random clips. So I challenged Seedance 2.5 on Pollo AI with something much harder—a 30-second Hollywood action trailer. Instead of disconnected scenes, it delivered a cinematic sequence with smooth shot transitions, realistic character performances, dynamic camera work, synchronized dialogue, immersive sound design, and blockbuster-style action. The biggest takeaway? When the model understands the story, the final video feels far more like a film than an AI generation.

Most AI videos still look like a collection of random clips. So I challenged Seedance 2.5 on Pollo AI with something much harder—a 30-second Hollywood action trailer. Instead of disconnected scenes, it delivered a cinematic sequence with smooth shot transitions, realistic character performances, dynamic camera work, synchronized dialogue, immersive sound design, and blockbuster-style action. The biggest takeaway? When the model understands the story, the final video feels far more like a film than an AI generation.

Shabnam Parveen

46,876 次观看 • 1 个月前

AI video is quietly shifting from generation to direction. After testing Ray 3.2 on a restaurant launch concept, that became pretty obvious. I started with two static images: one exterior shot, one interior dining scene. A few minutes later, I had a cinematic sequence that actually felt connected. The biggest improvement isn't just realism. It's control. Most AI video workflows still look like this: prompt → generate → regenerate → regenerate again. You're mostly hoping the model understands what you want. Ray 3.2 feels different. Instead of generating isolated clips, you can guide how one scene evolves into the next. You can shape transitions, control camera movement, preserve visual consistency, and build sequences that feel intentionally directed. For a simple restaurant commercial, that makes a huge difference. The exterior naturally flows into the dining experience. The pacing feels deliberate. The motion feels cinematic. The story feels connected. And that's what stood out to me most. For the first time, AI video felt less like prompting and more like directing. Every frame, yours to direct. Try Luma today using my link:

AI video is quietly shifting from generation to direction. After testing Ray 3.2 on a restaurant launch concept, that became pretty obvious. I started with two static images: one exterior shot, one interior dining scene. A few minutes later, I had a cinematic sequence that actually felt connected. The biggest improvement isn't just realism. It's control. Most AI video workflows still look like this: prompt → generate → regenerate → regenerate again. You're mostly hoping the model understands what you want. Ray 3.2 feels different. Instead of generating isolated clips, you can guide how one scene evolves into the next. You can shape transitions, control camera movement, preserve visual consistency, and build sequences that feel intentionally directed. For a simple restaurant commercial, that makes a huge difference. The exterior naturally flows into the dining experience. The pacing feels deliberate. The motion feels cinematic. The story feels connected. And that's what stood out to me most. For the first time, AI video felt less like prompting and more like directing. Every frame, yours to direct. Try Luma today using my link:

SONIA

41,651 次观看 • 3 个月前

$P-Video-Animate is now the fastest model for top-quality image-to-video motion transfer. - $0.03/s at 720p and $0.06/s at 1080p — a fraction of typical motion-transfer pricing (~$0.07–$0.35/s at 720p in our benchmarks) - 5.24 s generation time per second of output video at 720p — roughly 7× faster than typical motion-transfer alternatives in the same benchmarks - Upload one still plus a driver video to preserve motion, timing, camera movement, and scene structure, with optional source audio and clips up to ~2 minutes at 720p Ideal for UGC ads, gaming cinematics, film casting, and product demos when you already have a winning take and want a new hero still to follow it. Try it in the playground: We’re launching with 70% discounted usage for the coming two days with our launch partners. each::labs, Replicate, Runware, Scenario, WaveSpeedAI, Wiro, Tellers, Soon also on Together AI We’d love your feedback as you try it. 📚 Model docs & API examples: 📚 API quickstart: ⭐️ OSS: 🌐$

P-Video-Animate is now the fastest model for top-quality image-to-video motion transfer. - $0.03/s at 720p and $0.06/s at 1080p — a fraction of typical motion-transfer pricing (~$0.07–$0.35/s at 720p in our benchmarks) - 5.24 s generation time per second of output video at 720p — roughly 7× faster than typical motion-transfer alternatives in the same benchmarks - Upload one still plus a driver video to preserve motion, timing, camera movement, and scene structure, with optional source audio and clips up to ~2 minutes at 720p Ideal for UGC ads, gaming cinematics, film casting, and product demos when you already have a winning take and want a new hero still to follow it. Try it in the playground: We’re launching with 70% discounted usage for the coming two days with our launch partners. each::labs, Replicate, Runware, Scenario, WaveSpeedAI, Wiro, Tellers, Soon also on Together AI We’d love your feedback as you try it. 📚 Model docs & API examples: 📚 API quickstart: ⭐️ OSS: 🌐

Pruna AI

22,727 次观看 • 3 个月前

Today, we introduce the next evolution in AI video generation. Meet Viggle LIVE: generate video, capture motion, and swap characters — in real time! With just an image and your webcam, you can transform into anyone live. It opens the door to creative applications such as: - V-tubing and faceless live streaming - VFX previsualization - Joining meetings as an AI avatar - Virtual cosplay and AR effects To celebrate the launch, we're giving away one-month free access to Viggle LIVE. Comment who you'd turn into & repost to get a chance to win!👇

ViggleAI

67,289 次观看 • 1 年前

Turning video into humanoid robot motion! 🤳🏼 Training humanoid robots needs huge amounts of motion data, but real-world capture doesn’t scale. Mocap is expensive, dangerous edge cases are rare, and you can’t ask humans to repeatedly fall or crash. Video2Robot tackles this by converting videos into physics-grounded humanoid simulations. Motion is generated to respect balance, inertia, ground contact, and joint limits, then directly retargeted to robot simulators. One prompt can generate a full humanoid motion sequence, including multi-agent interactions and failure cases like falls or collisions, scenarios that are hard or impossible to capture safely in the real world. The pipeline is model-agnostic and works with existing video generators, making it a practical way to scale data for robots. If robots are going to operate in the real world, they need to be trained on the failures too, not just the perfect demos. Here's the GitHub: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Turning video into humanoid robot motion! 🤳🏼 Training humanoid robots needs huge amounts of motion data, but real-world capture doesn’t scale. Mocap is expensive, dangerous edge cases are rare, and you can’t ask humans to repeatedly fall or crash. Video2Robot tackles this by converting videos into physics-grounded humanoid simulations. Motion is generated to respect balance, inertia, ground contact, and joint limits, then directly retargeted to robot simulators. One prompt can generate a full humanoid motion sequence, including multi-agent interactions and failure cases like falls or collisions, scenarios that are hard or impossible to capture safely in the real world. The pipeline is model-agnostic and works with existing video generators, making it a practical way to scale data for robots. If robots are going to operate in the real world, they need to be trained on the failures too, not just the perfect demos. Here's the GitHub: ~~ ♻️ Join the weekly robotics newsletter, and never miss any news →

Lukas Ziegler

59,835 次观看 • 8 个月前

Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models paper page: Traditional 3D content creation tools empower users to bring their imagination to life by giving them direct control over a scene's geometry, appearance, motion, and camera path. Creating computer-generated videos, however, is a tedious manual process, which can be automated by emerging text-to-video diffusion models. Despite great promise, video diffusion models are difficult to control, hindering a user to apply their own creativity rather than amplifying it. To address this challenge, we present a novel approach that combines the controllability of dynamic 3D meshes with the expressivity and editability of emerging diffusion models. For this purpose, our approach takes an animated, low-fidelity rendered mesh as input and injects the ground truth correspondence information obtained from the dynamic mesh into various stages of a pre-trained text-to-image generation model to output high-quality and temporally consistent frames. We demonstrate our approach on various examples where motion can be obtained by animating rigged assets or changing the camera path.

Generative Rendering: Controllable 4D-Guided Video Generation with 2D Diffusion Models paper page: Traditional 3D content creation tools empower users to bring their imagination to life by giving them direct control over a scene's geometry, appearance, motion, and camera path. Creating computer-generated videos, however, is a tedious manual process, which can be automated by emerging text-to-video diffusion models. Despite great promise, video diffusion models are difficult to control, hindering a user to apply their own creativity rather than amplifying it. To address this challenge, we present a novel approach that combines the controllability of dynamic 3D meshes with the expressivity and editability of emerging diffusion models. For this purpose, our approach takes an animated, low-fidelity rendered mesh as input and injects the ground truth correspondence information obtained from the dynamic mesh into various stages of a pre-trained text-to-image generation model to output high-quality and temporally consistent frames. We demonstrate our approach on various examples where motion can be obtained by animating rigged assets or changing the camera path.

AK

117,068 次观看 • 2 年前