Загрузка видео...

Не удалось загрузить видео

На главную

🚀 Introducing #SparseVideoGen: 2x speedup in video generation with HunyuanVideo with high pixel-level fidelity (PSNR = 29)! No training is required, no perceptible difference to the human eye! Blog: Paper: Code: #ArtificialIntelligence #OpenSource [1/7]

HaochengXiUCB's profile picture

Haocheng Xi

1,514 subscribers

45,770 просмотров • 1 год назад •via X (Twitter)

#SparseVideoGen
LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

[1/N]🚀New decoding paradigm drop!🚀 Introducing Lookahead Reasoning(LR): step-level speculation that stacks with Speculative Decoding(SD). It has been accepted to #NeurIPS2025 🎉 📖 Blog: 💻 Code: 📄 Paper:
0:22

[1/N]🚀New decoding paradigm drop!🚀 Introducing Lookahead Reasoning(LR): step-level speculation that stacks with Speculative Decoding(SD). It has been accepted to #NeurIPS2025 🎉 📖 Blog: 💻 Code: 📄 Paper:

Hao AI Lab

43,162 просмотров • 9 месяцев назад

🎉🚀 Introducing Noya (YC W23) 🚀🎉 Noya is a wireframing tool that generates designs & code. Try it out: Noya lets you work in low-fidelity wireframes, while still producing high-fidelity designs and React prototypes to share with your team.
2:17

🎉🚀 Introducing Noya (YC W23) 🚀🎉 Noya is a wireframing tool that generates designs & code. Try it out: Noya lets you work in low-fidelity wireframes, while still producing high-fidelity designs and React prototypes to share with your team.

Devin Abbott

108,279 просмотров • 3 лет назад

🚀Turn Single Image into 3D Human🚀 #GeneMAN is a generalizable single-image 3D human reconstruction framework that turns in-the-wild images into high-quality 3D humans with ease 🔗Project: 📜Paper: 🧑‍💻Code:
1:04

🚀Turn Single Image into 3D Human🚀 #GeneMAN is a generalizable single-image 3D human reconstruction framework that turns in-the-wild images into high-quality 3D humans with ease 🔗Project: 📜Paper: 🧑‍💻Code:

Ziwei Liu

26,953 просмотров • 1 год назад

🚀 Introducing #DreamCraft3D, our breakthrough hierarchical technique for 3D content generation. Leading the way in AIGC, we're redefining 3D generation: Project page: Paper: Code:
1:34

🚀 Introducing #DreamCraft3D, our breakthrough hierarchical technique for 3D content generation. Leading the way in AIGC, we're redefining 3D generation: Project page: Paper: Code:

DeepSeek

16,400 просмотров • 2 лет назад

THE WAIT IS OVER 🚀 Introducing $SPCH - the Leverage Shares 2x Long SpaceX Daily ETF. Built for traders looking to amplify exposure to SpaceX. - 2x daily exposure - Trades like a stock - No margin account required Now live on Cboe
0:22

THE WAIT IS OVER 🚀 Introducing $SPCH - the Leverage Shares 2x Long SpaceX Daily ETF. Built for traders looking to amplify exposure to SpaceX. - 2x daily exposure - Trades like a stock - No margin account required Now live on Cboe

Leverage Shares ETFs

5,286,112 просмотров • 25 дней назад

🔥Introducing #𝐌𝐚𝐭𝐀𝐧𝐲𝐨𝐧𝐞 for human video matting!🔥 🤡 Fast video matting with customizable target 🤡 Stable human tracking 🤡 Fine-grained matting details - Paper: - Project: - Code (soon):
0:22

🔥Introducing #𝐌𝐚𝐭𝐀𝐧𝐲𝐨𝐧𝐞 for human video matting!🔥 🤡 Fast video matting with customizable target 🤡 Stable human tracking 🤡 Fine-grained matting details - Paper: - Project: - Code (soon):

Shangchen Zhou

42,487 просмотров • 1 год назад

The latent-vs-pixel debate misses the point. GPT Image 2 shows what users notice: pixel-level fidelity. Latent models show what scales: compact semantic structure. We connect them by replacing VAE/RAE decoders with a Pixel Diffusion Decoder. Code and Model available: 🧵(1/N)
0:48

The latent-vs-pixel debate misses the point. GPT Image 2 shows what users notice: pixel-level fidelity. Latent models show what scales: compact semantic structure. We connect them by replacing VAE/RAE decoders with a Pixel Diffusion Decoder. Code and Model available: 🧵(1/N)

Xuanchi Ren

669,031 просмотров • 1 месяц назад

Introducing LingBot-World: An open-source world simulator pushing the boundaries of video generation. 🚀 🌍 High-Fidelity: Realistic, scientific, & stylized. 🧠 Long-Term Memory: Minute-level consistency. ⚡ Real-Time: <1s latency at 16 FPS. 📜 Apache 2.0 Licensed. Model: Github:
0:29

Introducing LingBot-World: An open-source world simulator pushing the boundaries of video generation. 🚀 🌍 High-Fidelity: Realistic, scientific, & stylized. 🧠 Long-Term Memory: Minute-level consistency. ⚡ Real-Time: <1s latency at 16 FPS. 📜 Apache 2.0 Licensed. Model: Github:

ModelScope

28,809 просмотров • 5 месяцев назад

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.
1:00

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 просмотров • 1 год назад

Make Pixels Dance: High-Dynamic Video Generation paper page: Creating high-dynamic videos such as motion-rich actions and sophisticated visual effects poses a significant challenge in the field of artificial intelligence. Unfortunately, current state-of-the-art video generation methods, primarily focusing on text-to-video generation, tend to produce video clips with minimal motions despite maintaining high fidelity. We argue that relying solely on text instructions is insufficient and suboptimal for video generation. In this paper, we introduce PixelDance, a novel approach based on diffusion models that incorporates image instructions for both the first and last frames in conjunction with text instructions for video generation. Comprehensive experimental results demonstrate that PixelDance trained with public data exhibits significantly better proficiency in synthesizing videos with complex scenes and intricate motions, setting a new standard for video generation.
3:09

Make Pixels Dance: High-Dynamic Video Generation paper page: Creating high-dynamic videos such as motion-rich actions and sophisticated visual effects poses a significant challenge in the field of artificial intelligence. Unfortunately, current state-of-the-art video generation methods, primarily focusing on text-to-video generation, tend to produce video clips with minimal motions despite maintaining high fidelity. We argue that relying solely on text instructions is insufficient and suboptimal for video generation. In this paper, we introduce PixelDance, a novel approach based on diffusion models that incorporates image instructions for both the first and last frames in conjunction with text instructions for video generation. Comprehensive experimental results demonstrate that PixelDance trained with public data exhibits significantly better proficiency in synthesizing videos with complex scenes and intricate motions, setting a new standard for video generation.

AK

101,655 просмотров • 2 лет назад

Speeding your view synthesis(<40s) with #InstantSplat! Our large-scale, pose-free method trains in just 37 seconds from sparse views—no #COLMAP, no intrinsics needed. Achieving nearly 30dB test PSNR with just 12 images, New standard in #NVS and new training efficiency. Project page 👉 Paper 📷:
0:22

Speeding your view synthesis(<40s) with #InstantSplat! Our large-scale, pose-free method trains in just 37 seconds from sparse views—no #COLMAP, no intrinsics needed. Achieving nearly 30dB test PSNR with just 12 images, New standard in #NVS and new training efficiency. Project page 👉 Paper 📷:

Zhiwen(Aaron) Fan

108,763 просмотров • 2 лет назад

🎥 Videos DiTs are painfully slow, HunyuanVideo takes 16 min to generate a 5s 720P video on H100. 🤯 Announcing Sliding Tile Attention (STA): * Accelerate 3D full attention (FA3) by up to 10x * Slash the end-to-end time from 16 --> 5 mins * NO extra training. NO quality loss! 🚀 Can you tell which videos are generated by the original HunyuanVideo, and which by STA? 👀 Blog:
0:42

🎥 Videos DiTs are painfully slow, HunyuanVideo takes 16 min to generate a 5s 720P video on H100. 🤯 Announcing Sliding Tile Attention (STA): * Accelerate 3D full attention (FA3) by up to 10x * Slash the end-to-end time from 16 --> 5 mins * NO extra training. NO quality loss! 🚀 Can you tell which videos are generated by the original HunyuanVideo, and which by STA? 👀 Blog:

Hao AI Lab

58,025 просмотров • 1 год назад

Introducing quickarXiv Papers are often written in convoluted language that is hard to understand. We're fixing that. Swap arxiv → quickarxiv on any paper URL to get an instant blog with figures, insights, and explanations. Now extracted with DeepSeek OCR 🚀
0:31

Introducing quickarXiv Papers are often written in convoluted language that is hard to understand. We're fixing that. Swap arxiv → quickarxiv on any paper URL to get an instant blog with figures, insights, and explanations. Now extracted with DeepSeek OCR 🚀

alphaXiv

137,775 просмотров • 7 месяцев назад

Netflix Engineering: DB Migration to Aurora results in 75% speedup Blog: Paper: System Design Course: #Netflix #Aurora #DatabaseMigration
1:32

Netflix Engineering: DB Migration to Aurora results in 75% speedup Blog: Paper: System Design Course: #Netflix #Aurora #DatabaseMigration

Gaurav Sen

15,217 просмотров • 7 месяцев назад

introducing Crawlbyte scrape any website on the internet with no code 2x as fast as browser solutions we just changed the scraping industry (thread):
1:50

introducing Crawlbyte scrape any website on the internet with no code 2x as fast as browser solutions we just changed the scraping industry (thread):

Gunnar

335,781 просмотров • 8 месяцев назад

Nvidia and UC San Diego announce What You See is What You GAN Rendering Every Pixel for High-Fidelity Geometry in 3D GANs paper page: scale neural volume rendering to high resolution by rendering every pixel to ensure that "what you see in 2D, is what you get in 3D"
3:11

Nvidia and UC San Diego announce What You See is What You GAN Rendering Every Pixel for High-Fidelity Geometry in 3D GANs paper page: scale neural volume rendering to high resolution by rendering every pixel to ensure that "what you see in 2D, is what you get in 3D"

AK

127,419 просмотров • 2 лет назад

𝗢𝗽𝗲𝗻𝗔𝗿𝗺 is an open-source robot arm designed for human manipulation data collection. What's new in 𝘃𝟬.𝟮 (beta): ・Teleop with gravity compensation ・Bilateral force feedback support #OpenSource #Robotics #Humanoids #ArtificialIntelligence
0:39

𝗢𝗽𝗲𝗻𝗔𝗿𝗺 is an open-source robot arm designed for human manipulation data collection. What's new in 𝘃𝟬.𝟮 (beta): ・Teleop with gravity compensation ・Bilateral force feedback support #OpenSource #Robotics #Humanoids #ArtificialIntelligence

Reazon Human Interaction Lab

43,128 просмотров • 1 год назад

we built the world’s first self-installing API powered by claude code add memory to your AI agents in 5 mins with Hyperspell (YC F25) no human coding required
1:43

we built the world’s first self-installing API powered by claude code add memory to your AI agents in 5 mins with Hyperspell (YC F25) no human coding required

conor brennan-burke

29,355 просмотров • 8 месяцев назад

The code of GSPN #CVPR2025 is released! We proposed a new sqrt(N) complexity attention mechanism, which enables efficient high resolution image generation. We can generate 8k images with 42x speed up compared to self-attention in StableDiffusionXL! Code: Paper:
0:35

The code of GSPN #CVPR2025 is released! We proposed a new sqrt(N) complexity attention mechanism, which enables efficient high resolution image generation. We can generate 8k images with 42x speed up compared to self-attention in StableDiffusionXL! Code: Paper:

Xiaolong Wang

354,887 просмотров • 1 год назад