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。

Introducing StoryMem — a memory-augmented framework for multi-shot long video storytelling. StoryMem carefully injects compact memory into the generation process with minimal overhead, enabling: • Cross-shot consistency • Smooth transitions • Narrative coherence across minutes-long videos Awesome work by Kaiwen Kaiwen Zhang (Kevin) Project: arXiv: Code:

Xingang Pan

3,326 subscribers

16,474 次观看 • 6 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

Introducing Wan2.6 - A native multimodal model that turns your ideas into breathtaking videos and images! · Starring: Cast characters from reference videos into new scenes. Support human or human-like figures, enabling complex multi-person and human-object interactions with appearance and voice consistency. · Intelligent Multi-shot Narrative: Turn simple prompts into auto-storyboarded, multi-shot videos. Maintain visual consistency and upgrade storytelling from single shots to rich narratives. · Native A/V Sync: Generate multi-speaker dialogue with natural lip-sync and studio-quality audio. It doesn’t just look real - it sounds real. · Cinematic Quality: 15s 1080p HD generation with comprehensive upgrades to instruction adherence, motion physics, and aesthetic control. · Advanced Image Synthesis and Editing: Deliver cinematic photorealism with precise control over lens and lighting. Support multi-image referencing for commercial-grade consistency and faithful aesthetic transfer. · Storytelling with Structure: Generate interleaved texts and images powered by real-world knowledge and reasoning capabilities, enabling hierarchical and structured visual narratives.

Introducing Wan2.6 - A native multimodal model that turns your ideas into breathtaking videos and images! · Starring: Cast characters from reference videos into new scenes. Support human or human-like figures, enabling complex multi-person and human-object interactions with appearance and voice consistency. · Intelligent Multi-shot Narrative: Turn simple prompts into auto-storyboarded, multi-shot videos. Maintain visual consistency and upgrade storytelling from single shots to rich narratives. · Native A/V Sync: Generate multi-speaker dialogue with natural lip-sync and studio-quality audio. It doesn’t just look real - it sounds real. · Cinematic Quality: 15s 1080p HD generation with comprehensive upgrades to instruction adherence, motion physics, and aesthetic control. · Advanced Image Synthesis and Editing: Deliver cinematic photorealism with precise control over lens and lighting. Support multi-image referencing for commercial-grade consistency and faithful aesthetic transfer. · Storytelling with Structure: Generate interleaved texts and images powered by real-world knowledge and reasoning capabilities, enabling hierarchical and structured visual narratives.

Wan

3,847,477 次观看 • 7 个月前

Introducing The Matrix --- a foundation world model for generating infinite-length, hyper-realistic videos with real-time, frame-level control: - Infinite-length video generation - 720p high-quality rendering - Real-time, frame-level control at 16 FPS - Generalization to real-world video control 🔗Blog: 📄Paper: 💻Code & Playable Demo: Coming soon! Key Innovation: A brand new technique called the shift-window denoise process model, enabling auto-regressive generation for diffusion and consistency models in real-time. Special thanks to project leader Ruili Feng and the entire Matrix team for their dedication and hard work over the year-long project.

Hongyang Zhang

178,322 次观看 • 1 年前

🚀Seedance 2.0 is now open to EVERYONE, including US. Enjoy no queues & full power on Lovart. → Unmatched Length: 60-second video generation with Lovart Skills → Consistency Control: Auto three-view generation across long videos → Creative Prompts: Autonomous planning, no complex shot prompts needed → Smart Storyboarding: Seamless shot merging for cinematic transitions Plus, members enjoy 45% OFF + 150 FREE Seedance 2.0 Videos. Like + repost + follow – 10 lucky winners get a free month of Pro

🚀Seedance 2.0 is now open to EVERYONE, including US. Enjoy no queues & full power on Lovart. → Unmatched Length: 60-second video generation with Lovart Skills → Consistency Control: Auto three-view generation across long videos → Creative Prompts: Autonomous planning, no complex shot prompts needed → Smart Storyboarding: Seamless shot merging for cinematic transitions Plus, members enjoy 45% OFF + 150 FREE Seedance 2.0 Videos. Like + repost + follow – 10 lucky winners get a free month of Pro

LovartAI

1,682,360 次观看 • 3 个月前

New research with Tsinghua University: Spatial-TTT. A framework for streaming visual-based spatial intelligence with test-time training (TTT). Spatial-TTT adapts fast weights to capture and organize spatial evidence from long video streams, enabling models to build structured 3D spatial memory over time. Highlights: 🔹Efficient streaming memory. Fast weights act as compact spatial memory with sublinear memory growth over 7000+ frames and more than 40% lower compute. 🔹Spatial-predictive mechanism. TTT layers with 3D spatiotemporal convolution capture geometric correspondence and temporal continuity. 🔹SOTA results on long-horizon video spatial understanding (VSI-Bench). The paper ranked #1 on Hugging Face Daily Papers on March 13. Project page: GitHub: Paper: Model & Data:

New research with Tsinghua University: Spatial-TTT. A framework for streaming visual-based spatial intelligence with test-time training (TTT). Spatial-TTT adapts fast weights to capture and organize spatial evidence from long video streams, enabling models to build structured 3D spatial memory over time. Highlights: 🔹Efficient streaming memory. Fast weights act as compact spatial memory with sublinear memory growth over 7000+ frames and more than 40% lower compute. 🔹Spatial-predictive mechanism. TTT layers with 3D spatiotemporal convolution capture geometric correspondence and temporal continuity. 🔹SOTA results on long-horizon video spatial understanding (VSI-Bench). The paper ranked #1 on Hugging Face Daily Papers on March 13. Project page: GitHub: Paper: Model & Data:

Tencent Hy

20,792 次观看 • 4 个月前

Meet LongCat-Video-Avatar: a robust audio-driven avatar model that pushes the boundaries of long-form video generation. Compared with the previous InfiniteTalk, LongCat-Video-Avatar delivers far better long-sequence stability and realism. New highlights： ⚙ Built on the LongCat-Video architecture, now supporting Audio-Text-to-Video (AT2V), Audio-Text-Image-to-Video (ATI2V), and Video Continuation modes. 🎭 Open-source SOTA Realism: Ranked 1st in overall anthropomorphism scores for both single and multi-subject scenarios in EvalTalker evaluations (492 participants, 3 independent raters per video). ♾ High-quality long videos: Cross-Chunk Latent Stitching prevents pixel degradation and error accumulation over time, ensuring seamless stitching quality. 🔒 Long-term consistency: Reference Skip Attention maintains ID consistency while eliminating rigid copy-paste artifacts. 🪄 Supports multi-person and infinite-length video generation. 🔗Open-sourced Code: Hugging Face: Project: Paper:

Meituan LongCat

28,041 次观看 • 7 个月前

🚀🚀🚀Introducing HY World 1.5 (WorldPlay)! We have now open-sourced the most systemized, comprehensive real-time world model framework in the industry. In HY World 1.5, we develop WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. You can generate and explore 3D worlds simply by inputting text or images. Walk, look around, and interact like you're playing a game. Highlights: 🔹Real-Time: Generates long-horizon streaming video at 24 FPS with superior consistency. 🔹Geometric Consistency: Achieved using a Reconstituted Context Memory mechanism to dynamically rebuild context from past frames to alleviate memory attenuation 🔹Robust Control: Uses a Dual Action Representation for robust response to user keyboard and mouse inputs. 🔹Versatile Applications: Supports both first-person and third-person perspectives, enabling applications like promptable events and infinite world extension. 👉🏻Try it now: 🌐Project Page: 🔗Github: 🤗Hugging Face: 📄Technical Report:

🚀🚀🚀Introducing HY World 1.5 (WorldPlay)! We have now open-sourced the most systemized, comprehensive real-time world model framework in the industry. In HY World 1.5, we develop WorldPlay, a streaming video diffusion model that enables real-time, interactive world modeling with long-term geometric consistency, resolving the trade-off between speed and memory that limits current methods. You can generate and explore 3D worlds simply by inputting text or images. Walk, look around, and interact like you're playing a game. Highlights: 🔹Real-Time: Generates long-horizon streaming video at 24 FPS with superior consistency. 🔹Geometric Consistency: Achieved using a Reconstituted Context Memory mechanism to dynamically rebuild context from past frames to alleviate memory attenuation 🔹Robust Control: Uses a Dual Action Representation for robust response to user keyboard and mouse inputs. 🔹Versatile Applications: Supports both first-person and third-person perspectives, enabling applications like promptable events and infinite world extension. 👉🏻Try it now: 🌐Project Page: 🔗Github: 🤗Hugging Face: 📄Technical Report:

Tencent Hy

189,802 次观看 • 7 个月前

HunyuanWorld-Voyager is here and fully open-source! The world’s first ultra-long-range world model with native 3D reconstruction, redefining AI-driven spatial intelligence for VR, gaming, and simulations. ✅Direct 3D Output: Exports point cloud videos to 3D formats without tools like COLMAP, enabling instant 3D application use. ✅Innovative 3D Memory: Introduces a scalable world caching mechanism, ensuring geometric consistency across any camera trajectory. ✅Top-Ranked Performance: #1 on Stanford’s WorldScore, excelling in video generation and 3D reconstruction benchmarks.( Built on HunyuanWorld 1.0, Voyager blends video generation with 3D modeling, delivering camera-controlled, high-fidelity RGB-D sequences. Control scenes via keyboard or joystick for unmatched 3D consistency. Explore now: 🌐Project Page: 🔗GitHub: 🤗HuggingFace: 📝Technical Details:

HunyuanWorld-Voyager is here and fully open-source! The world’s first ultra-long-range world model with native 3D reconstruction, redefining AI-driven spatial intelligence for VR, gaming, and simulations. ✅Direct 3D Output: Exports point cloud videos to 3D formats without tools like COLMAP, enabling instant 3D application use. ✅Innovative 3D Memory: Introduces a scalable world caching mechanism, ensuring geometric consistency across any camera trajectory. ✅Top-Ranked Performance: #1 on Stanford’s WorldScore, excelling in video generation and 3D reconstruction benchmarks.( Built on HunyuanWorld 1.0, Voyager blends video generation with 3D modeling, delivering camera-controlled, high-fidelity RGB-D sequences. Control scenes via keyboard or joystick for unmatched 3D consistency. Explore now: 🌐Project Page: 🔗GitHub: 🤗HuggingFace: 📝Technical Details:

Tencent Hy

198,289 次观看 • 10 个月前

[SIGGRAPH '26] Anchored Temporal Gaussian Splatting for Long Volumetric Video Representation TL;DR: We present ATGS, a novel framework for volumetric video reconstruction that effectively handles long sequences and complex motions. By utilizing time-conditioned anchors and a temporal windowing strategy, ATGS enhances temporal coherence and scalability. Abstract (excerpt): Key insight is that explicitly tracking long term complex motion with individual Gaussian primitives is inherently unstable. Instead, we organize Gaussians around time conditioned anchors that localize their spatial and temporal support, thereby reducing long range motion complexity. We further introduce a temporal windowing strategy to activate only anchors relevant to the queried time, which improves scalability and temporal coherence. In addition, to ensure spatial and temporal stability, we design a compact set of multi level anchor features that encode global features, local spatial features, and local temporal features, jointly constraining Gaussian generation. Extensive experiments demonstrate that ATGS consistently outperforms prior methods on long sequence volumetric videos with complex motions.

[SIGGRAPH '26] Anchored Temporal Gaussian Splatting for Long Volumetric Video Representation TL;DR: We present ATGS, a novel framework for volumetric video reconstruction that effectively handles long sequences and complex motions. By utilizing time-conditioned anchors and a temporal windowing strategy, ATGS enhances temporal coherence and scalability. Abstract (excerpt): Key insight is that explicitly tracking long term complex motion with individual Gaussian primitives is inherently unstable. Instead, we organize Gaussians around time conditioned anchors that localize their spatial and temporal support, thereby reducing long range motion complexity. We further introduce a temporal windowing strategy to activate only anchors relevant to the queried time, which improves scalability and temporal coherence. In addition, to ensure spatial and temporal stability, we design a compact set of multi level anchor features that encode global features, local spatial features, and local temporal features, jointly constraining Gaussian generation. Extensive experiments demonstrate that ATGS consistently outperforms prior methods on long sequence volumetric videos with complex motions.

MrNeRF

26,905 次观看 • 3 个月前

We analyzed Anthropic’s memory architecture and built something better: a persistent, human-inspectable, and token-efficient memory layer that scales with your projects. Today, it’s OPEN-SOURCE. ByteRover CLI gives agents (like OpenClaw, Claude Code, and Hermes) persistent, structured memory. Built on the exact architecture that became the #1 memory system for OpenClaw (30,000+ downloads in a week), it lets developers curate project knowledge into a file-based hierarchy. This guarantees highly accurate, lightning-fast retrieval, even with lightweight models. 👉Highly Accurate: >92% retrieval accuracy across long-running sessions - the highest proven production accuracy on the market. ⚡Fast: ~1.6s average retrieval time. 💰Economic: Maintains >90% accuracy even with lightweight models, saving 50-70% on token costs. ☁ Portable: Runs locally by default, with cloud-sync to share memory across agents and teammates.

We analyzed Anthropic’s memory architecture and built something better: a persistent, human-inspectable, and token-efficient memory layer that scales with your projects. Today, it’s OPEN-SOURCE. ByteRover CLI gives agents (like OpenClaw, Claude Code, and Hermes) persistent, structured memory. Built on the exact architecture that became the #1 memory system for OpenClaw (30,000+ downloads in a week), it lets developers curate project knowledge into a file-based hierarchy. This guarantees highly accurate, lightning-fast retrieval, even with lightweight models. 👉Highly Accurate: >92% retrieval accuracy across long-running sessions - the highest proven production accuracy on the market. ⚡Fast: ~1.6s average retrieval time. 💰Economic: Maintains >90% accuracy even with lightweight models, saving 50-70% on token costs. ☁ Portable: Runs locally by default, with cloud-sync to share memory across agents and teammates.

andy nguyen

28,980 次观看 • 3 个月前

We’re introducing Zopia — an end-to-end AI Video Director Agent for Film-grade creation. One idea + Zopia = a cinematic blockbuster. Zopia intelligently orchestrates leading models, enabling one-stop creation of AI Short Dramas, Films, Branded videos and more.. Through conversational multi-Agent + Skills coordination, you bring a story — Zopia delivers the finished film. An industry-level filmmaking pipeline: Script → Storyboard → Multi-shot Generation → Final Edit. Zopia solves the hardest problems in AI filmmaking: shot continuity, character consistency, and multi-scene alignment. Fully optimized for Kling 3 and Vidu Q3, Zopia produces realistic live-action scenes and coherent cinematic storyboards. Seedance 2.0 integration is coming soon — unlocking even greater expressive power. Zopia can be triggered by OpenClaw and other Bots for 24/7 automated production. You imagine. We produce.

We’re introducing Zopia — an end-to-end AI Video Director Agent for Film-grade creation. One idea + Zopia = a cinematic blockbuster. Zopia intelligently orchestrates leading models, enabling one-stop creation of AI Short Dramas, Films, Branded videos and more.. Through conversational multi-Agent + Skills coordination, you bring a story — Zopia delivers the finished film. An industry-level filmmaking pipeline: Script → Storyboard → Multi-shot Generation → Final Edit. Zopia solves the hardest problems in AI filmmaking: shot continuity, character consistency, and multi-scene alignment. Fully optimized for Kling 3 and Vidu Q3, Zopia produces realistic live-action scenes and coherent cinematic storyboards. Seedance 2.0 integration is coming soon — unlocking even greater expressive power. Zopia can be triggered by OpenClaw and other Bots for 24/7 automated production. You imagine. We produce.

Zopia AI

294,412 次观看 • 4 个月前

Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution paper page: Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.

Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution paper page: Text-based diffusion models have exhibited remarkable success in generation and editing, showing great promise for enhancing visual content with their generative prior. However, applying these models to video super-resolution remains challenging due to the high demands for output fidelity and temporal consistency, which is complicated by the inherent randomness in diffusion models. Our study introduces Upscale-A-Video, a text-guided latent diffusion framework for video upscaling. This framework ensures temporal coherence through two key mechanisms: locally, it integrates temporal layers into U-Net and VAE-Decoder, maintaining consistency within short sequences; globally, without training, a flow-guided recurrent latent propagation module is introduced to enhance overall video stability by propagating and fusing latent across the entire sequences. Thanks to the diffusion paradigm, our model also offers greater flexibility by allowing text prompts to guide texture creation and adjustable noise levels to balance restoration and generation, enabling a trade-off between fidelity and quality. Extensive experiments show that Upscale-A-Video surpasses existing methods in both synthetic and real-world benchmarks, as well as in AI-generated videos, showcasing impressive visual realism and temporal consistency.

AK

32,849 次观看 • 2 年前

Co-produced with PAI by Utopai Studios, the first long-form video generation model. The AI video clip era is ending! I started with a short story idea. From there, PAI generated the characters, transformed them into keyframes, and then animated the scenes into a full sequence. The interesting part was seeing how the same characters, environment, and visual style stayed consistent across multiple shots while the story progressed naturally. What felt different here is that I wasn’t building each shot manually. I focused on the narrative, and the engine handled character creation, scene continuity, and animation to produce a ready-to-watch video. A step-by-step tutorial so you can make your own story 👇🏻 Thanks to Utopai Studios for collaborating with me Try it now: #UtopaiPartners

Co-produced with PAI by Utopai Studios, the first long-form video generation model. The AI video clip era is ending! I started with a short story idea. From there, PAI generated the characters, transformed them into keyframes, and then animated the scenes into a full sequence. The interesting part was seeing how the same characters, environment, and visual style stayed consistent across multiple shots while the story progressed naturally. What felt different here is that I wasn’t building each shot manually. I focused on the narrative, and the engine handled character creation, scene continuity, and animation to produce a ready-to-watch video. A step-by-step tutorial so you can make your own story 👇🏻 Thanks to Utopai Studios for collaborating with me Try it now: #UtopaiPartners

Amira Zairi

25,679 次观看 • 4 个月前

I tested Wizstar by creating a cinematic smartphone promo to see how well it could handle commercial-style product videos. And honestly… the result looked surprisingly close to a real ad. What stood out most was how well it handled: → cinematic camera movement → gaming visuals → smooth scene transitions → product consistency across every shot No complex editing. No huge production setup. Just concept → AI-generated commercial. This feels especially powerful for e-commerce brands pushing short-form ads at scale. AI video tools are moving way beyond simple “content generation.” They’re starting to handle much larger parts of the creative workflow. Try it here:

I tested Wizstar by creating a cinematic smartphone promo to see how well it could handle commercial-style product videos. And honestly… the result looked surprisingly close to a real ad. What stood out most was how well it handled: → cinematic camera movement → gaming visuals → smooth scene transitions → product consistency across every shot No complex editing. No huge production setup. Just concept → AI-generated commercial. This feels especially powerful for e-commerce brands pushing short-form ads at scale. AI video tools are moving way beyond simple “content generation.” They’re starting to handle much larger parts of the creative workflow. Try it here:

RAVI KUMAR SAHU

60,601 次观看 • 2 个月前

🚀 Introducing HunyuanCustom: An open-source, multimodal-driven architecture for customized video generation, powered by HunyuanVideo-13B. Outperforming existing open-source models, it rivals top closed-source solutions! 🎥 Highlights: ✅Subject Consistency: Maintains identity across single & multi-subject video generation. ✅Multimodal Inputs: Supports text, images, audio, and video for highly controlled outputs. ✅High-Quality Output: Industry-leading detail, motion smoothness, and lighting realism. Key Features: 1️⃣Single-Subject Video: Upload an image + text (e.g., “He’s walking a dog”) to create coherent videos with new actions, outfits, and scenes. 2️⃣Multi-Subject Video: Generate videos with multiple subjects (e.g., a man drinking coffee in a cozy room) from separate image inputs. 3️⃣Audio-Driven Video: Sync audio with visuals for talking or singing in any scene—perfect for digital avatars, virtual customer service, and more. 4️⃣Video-Driven Video: Seamlessly insert or replace subjects into any video for creative enhancements. Dive Deeper: 🌐Project: 📝Technical Details: 💻Code: The single-subject video capability is open-sourced and live on the Hunyuan website, with more features to be released soon! 👉 Try it now: #HunyuanCustom #VideoGeneration #AI #TencentHunyuan

🚀 Introducing HunyuanCustom: An open-source, multimodal-driven architecture for customized video generation, powered by HunyuanVideo-13B. Outperforming existing open-source models, it rivals top closed-source solutions! 🎥 Highlights: ✅Subject Consistency: Maintains identity across single & multi-subject video generation. ✅Multimodal Inputs: Supports text, images, audio, and video for highly controlled outputs. ✅High-Quality Output: Industry-leading detail, motion smoothness, and lighting realism. Key Features: 1️⃣Single-Subject Video: Upload an image + text (e.g., “He’s walking a dog”) to create coherent videos with new actions, outfits, and scenes. 2️⃣Multi-Subject Video: Generate videos with multiple subjects (e.g., a man drinking coffee in a cozy room) from separate image inputs. 3️⃣Audio-Driven Video: Sync audio with visuals for talking or singing in any scene—perfect for digital avatars, virtual customer service, and more. 4️⃣Video-Driven Video: Seamlessly insert or replace subjects into any video for creative enhancements. Dive Deeper: 🌐Project: 📝Technical Details: 💻Code: The single-subject video capability is open-sourced and live on the Hunyuan website, with more features to be released soon! 👉 Try it now: #HunyuanCustom #VideoGeneration #AI #TencentHunyuan

Tencent HY

191,908 次观看 • 1 年前

MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model with Gradio demo local demo: This paper studies the human image animation task, which aims to generate a video of a certain reference identity following a particular motion sequence. Existing animation works typically employ the frame-warping technique to animate the reference image towards the target motion. Despite achieving reasonable results, these approaches face challenges in maintaining temporal consistency throughout the animation due to the lack of temporal modeling and poor preservation of reference identity. In this work, we introduce MagicAnimate, a diffusion-based framework that aims at enhancing temporal consistency, preserving reference image faithfully, and improving animation fidelity. To achieve this, we first develop a video diffusion model to encode temporal information. Second, to maintain the appearance coherence across frames, we introduce a novel appearance encoder to retain the intricate details of the reference image. Leveraging these two innovations, we further employ a simple video fusion technique to encourage smooth transitions for long video animation. Empirical results demonstrate the superiority of our method over baseline approaches on two benchmarks. Notably, our approach outperforms the strongest baseline by over 38% in terms of video fidelity on the challenging TikTok dancing dataset. Code and model will be made available.

MagicAnimate: Temporally Consistent Human Image Animation using Diffusion Model with Gradio demo local demo: This paper studies the human image animation task, which aims to generate a video of a certain reference identity following a particular motion sequence. Existing animation works typically employ the frame-warping technique to animate the reference image towards the target motion. Despite achieving reasonable results, these approaches face challenges in maintaining temporal consistency throughout the animation due to the lack of temporal modeling and poor preservation of reference identity. In this work, we introduce MagicAnimate, a diffusion-based framework that aims at enhancing temporal consistency, preserving reference image faithfully, and improving animation fidelity. To achieve this, we first develop a video diffusion model to encode temporal information. Second, to maintain the appearance coherence across frames, we introduce a novel appearance encoder to retain the intricate details of the reference image. Leveraging these two innovations, we further employ a simple video fusion technique to encourage smooth transitions for long video animation. Empirical results demonstrate the superiority of our method over baseline approaches on two benchmarks. Notably, our approach outperforms the strongest baseline by over 38% in terms of video fidelity on the challenging TikTok dancing dataset. Code and model will be made available.

AK

810,578 次观看 • 2 年前

I’m thrilled to announce that we just released GraspGen, a multi-year project we have been cooking at NVIDIA Robotics 🚀 GraspGen: A Diffusion-Based Framework for 6-DOF Grasping Grasping is a foundational challenge in robotics 🤖 — whether for industrial picking or general-purpose humanoids. VLA + real data collection is all the rage now but is expensive and scales poorly for this task. For every new gripper and/or scene, you’ll have to recollect the dataset in this paradigm for the best perf. 💡Key Idea: Since grasping is such a well-defined task in simulation - why can’t we just scale synthetic data generation and train a generative model for grasping? By embracing modularity and standardized grasp formats, we can make this a turnkey technology that works zero-shot for multiple settings. GraspGen is a modular framework for diffusion-based 6-DOF grasp generation that scales across embodiment types, observability conditions, clutter, task complexity. Key Features: ✅ Multi-embodiment support: suction, parallel-jaw, and multi-fingered grippers ✅ Generalization to partial + complete 3D point clouds ✅ Generalization to single-objects + cluttered scenes ✅ Modular design uses other robotics modules and foundation models (SAM2, cuRobo, FoundationStereo, FoundationPose). This allows GraspGen to focus on only one thing - grasp generation ✅ Training recipe: grasp discriminator is trained with On-Generator data from the diffusion model - so that it learns to correct the mistakes (if any) of the diffusion generator ✅ Real-time performance (~20 Hz) before any GPU acceleration; low memory footprint 📊 Results: • SOTA on the FetchBench [Han et al. CoRL 2024] benchmark • Zero-shot sim-to-real transfer on unknown objects and cluttered scenes • Dataset of 53M simulated grasps across 8K objects from Objaverse 📄 arXiv: 🌐 Website: 💻 Code: A huge thank you to everyone involved in this journey — excited to see what the community builds on top of it! Joint work with Clemens Eppner , Balakumar Sundaralingam , Yu-Wei, Jun Yamada Wentao Yuan and other collaborators #robotics #diffusionmodels #physicalAI #simtoreal

I’m thrilled to announce that we just released GraspGen, a multi-year project we have been cooking at NVIDIA Robotics 🚀 GraspGen: A Diffusion-Based Framework for 6-DOF Grasping Grasping is a foundational challenge in robotics 🤖 — whether for industrial picking or general-purpose humanoids. VLA + real data collection is all the rage now but is expensive and scales poorly for this task. For every new gripper and/or scene, you’ll have to recollect the dataset in this paradigm for the best perf. 💡Key Idea: Since grasping is such a well-defined task in simulation - why can’t we just scale synthetic data generation and train a generative model for grasping? By embracing modularity and standardized grasp formats, we can make this a turnkey technology that works zero-shot for multiple settings. GraspGen is a modular framework for diffusion-based 6-DOF grasp generation that scales across embodiment types, observability conditions, clutter, task complexity. Key Features: ✅ Multi-embodiment support: suction, parallel-jaw, and multi-fingered grippers ✅ Generalization to partial + complete 3D point clouds ✅ Generalization to single-objects + cluttered scenes ✅ Modular design uses other robotics modules and foundation models (SAM2, cuRobo, FoundationStereo, FoundationPose). This allows GraspGen to focus on only one thing - grasp generation ✅ Training recipe: grasp discriminator is trained with On-Generator data from the diffusion model - so that it learns to correct the mistakes (if any) of the diffusion generator ✅ Real-time performance (~20 Hz) before any GPU acceleration; low memory footprint 📊 Results: • SOTA on the FetchBench [Han et al. CoRL 2024] benchmark • Zero-shot sim-to-real transfer on unknown objects and cluttered scenes • Dataset of 53M simulated grasps across 8K objects from Objaverse 📄 arXiv: 🌐 Website: 💻 Code: A huge thank you to everyone involved in this journey — excited to see what the community builds on top of it! Joint work with Clemens Eppner , Balakumar Sundaralingam , Yu-Wei, Jun Yamada Wentao Yuan and other collaborators #robotics #diffusionmodels #physicalAI #simtoreal

Adithya Murali

23,841 次观看 • 1 年前

“Paying homage to Adobe’s classic ad—the one where creativity lights up the world. I recreated that vibe in just 5 hours with Kling O1. Adobe’s inspiration still shines, but Kling O1 makes creation jump from slow to effortless. Character consistency is rock-solid: based on a reference video, expressions, poses, and style stay aligned across shots. Editing is straightforward—import, select, remove, swap, or tweak with real-time previews. Perfect for style replacements, green-screen work, multi-angle generation, and turning a single image into a dynamic shot sequence. Bottom line: Kling O1 is fast, reliable, and creator-ready. Give it a spin. 🚀” — Kling AI Creative Partner @ BOB

“Paying homage to Adobe’s classic ad—the one where creativity lights up the world. I recreated that vibe in just 5 hours with Kling O1. Adobe’s inspiration still shines, but Kling O1 makes creation jump from slow to effortless. Character consistency is rock-solid: based on a reference video, expressions, poses, and style stay aligned across shots. Editing is straightforward—import, select, remove, swap, or tweak with real-time previews. Perfect for style replacements, green-screen work, multi-angle generation, and turning a single image into a dynamic shot sequence. Bottom line: Kling O1 is fast, reliable, and creator-ready. Give it a spin. 🚀” — Kling AI Creative Partner @ BOB

Kling AI

65,391 次观看 • 7 个月前

I just built an open NotebookLM clone! Here's what it can do for you: - Process multi-modal data - Scrape websites and YouTube videos - Create a unified knowledge base - Lets you do RAG over it - Remember every conversation - Generate a podcast 🎙️ The idea here is not to reinvent the wheel but to understand how one of the most powerful tools for learning and research actually works, by building it step-by-step! So by the end of this video, you'll learn how to: ↳ Process multimodal data (text, audio, video, URLs, and YouTube videos) into a format ready for LLMs ↳ Store everything in a vector database for fast retrieval ↳ Add a memory layer that remembers conversations and preferences for a personalized experience ↳ Chat with your knowledge base or generate podcasts using a fully open-source, locally running text-to-speech model The podcast generation feature is my favorite part! There's something powerful about turning written content into conversational audio that you can listen to while doing something else. The entire code is 100% open-source. I've shared a link in the replies! ____ Don't forget to drop a like if you enjoy my videos. It shows me I should be making more content like this. Cheers! :)

I just built an open NotebookLM clone! Here's what it can do for you: - Process multi-modal data - Scrape websites and YouTube videos - Create a unified knowledge base - Lets you do RAG over it - Remember every conversation - Generate a podcast 🎙️ The idea here is not to reinvent the wheel but to understand how one of the most powerful tools for learning and research actually works, by building it step-by-step! So by the end of this video, you'll learn how to: ↳ Process multimodal data (text, audio, video, URLs, and YouTube videos) into a format ready for LLMs ↳ Store everything in a vector database for fast retrieval ↳ Add a memory layer that remembers conversations and preferences for a personalized experience ↳ Chat with your knowledge base or generate podcasts using a fully open-source, locally running text-to-speech model The podcast generation feature is my favorite part! There's something powerful about turning written content into conversational audio that you can listen to while doing something else. The entire code is 100% open-source. I've shared a link in the replies! ____ Don't forget to drop a like if you enjoy my videos. It shows me I should be making more content like this. Cheers! :)

Akshay 🚀

83,187 次观看 • 9 个月前

🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives. Key Advantages: 🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow. 🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement. 🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses inference steps, boosting speed and lowering costs. This allows the quantized 13B model to run on consumer-grade GPUs like the RTX 4090. Project Page: Code: Technical Report: Hugging Face：

🚀We are thrilled to open-source Hunyuan-GameCraft, a high-dynamic interactive game video generation framework built on HunyuanVideo. It generates playable and physically realistic videos from a single scene image and user action signals, empowering creators and developers to "direct" games with first-person or third-person perspectives. Key Advantages: 🔹High Dynamics: Unifies standard keyboard inputs into a shared continuous action space, enabling high-precision control over velocity and angle. This allows for the exploration of complex trajectories, overcoming the stiff, limited motion of traditional models. It can also generate dynamic environmental content like moving clouds, rain, snow, and water flow. 🔹Long-term Consistency: Uses hybrid history condition to preserve the original scene information after significant movement. 🔹Significant Cost Reduction: No need for expensive modeling/rendering. PCM distillation compresses inference steps, boosting speed and lowering costs. This allows the quantized 13B model to run on consumer-grade GPUs like the RTX 4090. Project Page: Code: Technical Report: Hugging Face：

Tencent Hy

169,638 次观看 • 11 个月前