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

Shangchen Zhou

1,416 subscribers

42,487 次观看 • 1 年前 •via X (Twitter)

科学技术 #𝐌𝐚𝐭𝐀𝐧𝐲𝐨𝐧𝐞

Anya Rossi• Live Now

Private livecam show

10 条评论

Shangchen Zhou 的头像

Shangchen Zhou1 年前

This work is led by @peiqing001, and huge thanks to @ccloy and all collaborators!

Best Wallet 的头像

Best Wallet1 年前

🚀 Best Wallet - The ultimate 2025 #crypto wallet 💸

lillyguisnet 的头像

lillyguisnet1 年前

The hair 👑👏👏👏

lee 的头像

lee1 年前

Aweseome work!! if i understand it reencodes and passes the previous matte back through. Does it also do the foreground background color seperation similar to RVM or (originally introduced in FBA matting)? I think that helps a little to forecast the foreground and background colors as well because it also needs to learn the boundary to do that

Adarsh Baghel 的头像

Adarsh Baghel1 年前

We need a demo to test it out 🔥

Shangchen Zhou 的头像

Shangchen Zhou1 年前

Please stay tuned, the demo will drop around next week🚀

Redmond Hosting 的头像

Redmond Hosting1 年前

This is incredible! Can I ask about the licensing and when the code will be released? I might be interested in buying a commercial license if one exists.

Shangchen Zhou 的头像

Shangchen Zhou1 年前

Thanks! We are planning to release the code and demo sometime next week. This project is under the S-Lab license (refer to which restricts commercial use. If you are interested, feel free to reach out, and we can discuss further details.

kfant 的头像

kfant1 年前

code sooon 🤡😄

llmstock.com mc 的头像

llmstock.com mc1 年前

is there a live demo?

相关视频

Don’t ever hurt a man’s ego I repeat, never ever 🔥🔥🔥🔥🔥 Video is for attention… video lo vadiki paina statement ki em sambandam ledu 🤡🤡

Don’t ever hurt a man’s ego I repeat, never ever 🔥🔥🔥🔥🔥 Video is for attention… video lo vadiki paina statement ki em sambandam ledu 🤡🤡

యువసామ్రాట్ 🏃🏻‍➡️🏃🏻‍♀️

25,379 次观看 • 4 个月前

📍Reality Check for Pakistan 🤡 — Bhikaristan shivering after watching this video 🔥

📍Reality Check for Pakistan 🤡 — Bhikaristan shivering after watching this video 🔥

Megh Updates 🚨™

56,371 次观看 • 7 个月前

🚀 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]

🚀 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]

Haocheng Xi

45,770 次观看 • 1 年前

🤡🤡🤡🇷🇺😂😂😂 “They got us out — literally carrying us in their arms. They treated us gently, like precious vessels. It’s true! If I hadn’t gone through this myself, maybe I wouldn’t have believed it.” Determined to prove that they ‘liberated’ Kostiantynivka, the Russian MoD slapped together a staged propaganda video featuring supposed locals with distinctly Russian accents. The Russian babushkas clearly reciting a script praised Russian soldiers for rescuing them from Azov fighters, who were using them as human shields.😱

🤡🤡🤡🇷🇺😂😂😂 “They got us out — literally carrying us in their arms. They treated us gently, like precious vessels. It’s true! If I hadn’t gone through this myself, maybe I wouldn’t have believed it.” Determined to prove that they ‘liberated’ Kostiantynivka, the Russian MoD slapped together a staged propaganda video featuring supposed locals with distinctly Russian accents. The Russian babushkas clearly reciting a script praised Russian soldiers for rescuing them from Azov fighters, who were using them as human shields.😱

Natalka

28,884 次观看 • 1 个月前

⚡️🇺🇦That's a new one.... Foreign agent Dmitry Chernyshovsky stated that Russia deliberately does NOT target civilians in Ukraine, because this demonstrates the state's cold indifference to human life, which is why there is no extermination of Ukrainians. That is, Russia does not target civilians and does not engage in genocide, because Russia doesn't care about human lives. 👉No, he's not joking and this is not a AI video....🤡

⚡️🇺🇦That's a new one.... Foreign agent Dmitry Chernyshovsky stated that Russia deliberately does NOT target civilians in Ukraine, because this demonstrates the state's cold indifference to human life, which is why there is no extermination of Ukrainians. That is, Russia does not target civilians and does not engage in genocide, because Russia doesn't care about human lives. 👉No, he's not joking and this is not a AI video....🤡

SIMPLICIUS Ѱ

29,041 次观看 • 7 个月前

🔥 Introducing MVLift: Generate realistic 3D motion without any 3D training data - just using 2D poses from monocular videos! Applicable to human motion, human-object interaction & animal motion. Joint work w/ Jiajun Wu & Karen 💡 How? We reformulate 3D motion estimation as generating consistent multi-view 2D pose sequences. Our framework uses 2D motion diffusion to progressively establish multi-view consistency, requiring only single-view 2D pose sequences for training. Project: Video with demonstration: Paper:

🔥 Introducing MVLift: Generate realistic 3D motion without any 3D training data - just using 2D poses from monocular videos! Applicable to human motion, human-object interaction & animal motion. Joint work w/ Jiajun Wu & Karen 💡 How? We reformulate 3D motion estimation as generating consistent multi-view 2D pose sequences. Our framework uses 2D motion diffusion to progressively establish multi-view consistency, requiring only single-view 2D pose sequences for training. Project: Video with demonstration: Paper:

Jiaman Li

15,886 次观看 • 1 年前

Speaking of political extremism, YouTube has yet to take down this video wherein Asmongold not only calls for leftist public figures to be assassinated, but with zero self-awareness he then labels *himself* an "acceptable target" for murder. 🐀🤡

Speaking of political extremism, YouTube has yet to take down this video wherein Asmongold not only calls for leftist public figures to be assassinated, but with zero self-awareness he then labels himself an "acceptable target" for murder. 🐀🤡

gato fumante

313,361 次观看 • 11 个月前

📢 A new Once Human PVP experience is coming your way! 🔥 Brand-new series, Once Human: RaidZone is dropping soon! No safe zones, no mercy. Land and fight instantly! Only the bold will rise to claim the crown. Are you ready to dominate the battlefield? 🔎 Check out the video for all the details! 🔔Follow Once Human: RaidZone, the official account for all RaidZone updates! #OnceHuman #oncehumanmobile

Once Human

80,948 次观看 • 1 年前

🔥Ultra-Long Video World Model up to 5min🔥 ✨ We introduce #LongVie2, an end-to-end autoregressive video world model that supports continuous video generation lasting up to 5min with: 🕹️ Strong Controllability 📷 Long-term Visual Fidelity 🔒 Temporal Consistency - Project: - Code: - Paper: . Thanks to AK !

🔥Ultra-Long Video World Model up to 5min🔥 ✨ We introduce #LongVie2, an end-to-end autoregressive video world model that supports continuous video generation lasting up to 5min with: 🕹️ Strong Controllability 📷 Long-term Visual Fidelity 🔒 Temporal Consistency - Project: - Code: - Paper: . Thanks to AK !

Ziwei Liu

82,700 次观看 • 8 个月前

"POSCO against Hindu saint." 🤡🔥 Self-styled godman Mallikarjun Mutya from Karnataka, was charged under POCSO after a video showed him behaving inappropriately with a 7 year-old child girl. Disgusting! Save your daughters from such Babas.

"POSCO against Hindu saint." 🤡🔥 Self-styled godman Mallikarjun Mutya from Karnataka, was charged under POCSO after a video showed him behaving inappropriately with a 7 year-old child girl. Disgusting! Save your daughters from such Babas.

Suraj Kumar Bauddh

151,331 次观看 • 6 个月前

VIDEO REPLY 🎬 TO MUTHU 🤡💀🚶🚶🏼‍♂️ BY #Azeem & #Archana FANS 🔥💥 My fav 💙 Watch with earphones 🎧 Till the end 😉 Archana Ravichandran MOHAMED AZEEM #AzeemArmy #Soundarya #Soundariya #ArchanaArmy #BiggBossTamil #BiggBossTamil8 #ArchanaRavichandran

VIDEO REPLY 🎬 TO MUTHU 🤡💀🚶🚶🏼‍♂️ BY #Azeem & #Archana FANS 🔥💥 My fav 💙 Watch with earphones 🎧 Till the end 😉 Archana Ravichandran MOHAMED AZEEM #AzeemArmy #Soundarya #Soundariya #ArchanaArmy #BiggBossTamil #BiggBossTamil8 #ArchanaRavichandran

Sakthivel

12,284 次观看 • 1 年前

Humanoids finally move like humans… and can do more than copy. [Details + demos in thread 👇] A new framework, BeyondMimic, shows how to learn naturalistic whole-body control from human motion. But then goes further by composing those skills into versatile, zero-shot policies. Two key pieces make it work: •Motion tracking pipeline: robustly turns human mocap into dynamic robot motions (jump spins, sprinting, cartwheels) on real hardware •Guided diffusion policy: distills motion primitives and enables zero-shot downstream tasks with simple cost functions The result: ✅ State-of-the-art motion quality from mocap tracking ✅ Versatile task execution; waypoint navigation, joystick teleop, obstacle avoidance ✅ Stable sim-to-real transfer for diverse humanoid skills Thank you for sharing, Qiayuan Liao! 📍 Paper: Video:

Humanoids finally move like humans… and can do more than copy. [Details + demos in thread 👇] A new framework, BeyondMimic, shows how to learn naturalistic whole-body control from human motion. But then goes further by composing those skills into versatile, zero-shot policies. Two key pieces make it work: •Motion tracking pipeline: robustly turns human mocap into dynamic robot motions (jump spins, sprinting, cartwheels) on real hardware •Guided diffusion policy: distills motion primitives and enables zero-shot downstream tasks with simple cost functions The result: ✅ State-of-the-art motion quality from mocap tracking ✅ Versatile task execution; waypoint navigation, joystick teleop, obstacle avoidance ✅ Stable sim-to-real transfer for diverse humanoid skills Thank you for sharing, Qiayuan Liao! 📍 Paper: Video:

Ilir Aliu - eu/acc

72,522 次观看 • 1 年前

Video Game Fables is now available for the new low price of $13.99! Happy 1 Year Releasaversary ♥ 🤡Funny and lighthearted ⚔Fast-paced turn based combat with lots of twists 😈Crazy bosses that break the battle system 🎣Cozy stuff #indiegame #rpg #jrpg

Video Game Fables is now available for the new low price of $13.99! Happy 1 Year Releasaversary ♥ 🤡Funny and lighthearted ⚔Fast-paced turn based combat with lots of twists 😈Crazy bosses that break the battle system 🎣Cozy stuff #indiegame #rpg #jrpg

VIDEO GAME FABLES available now on Steam!

17,579 次观看 • 3 年前

🚀 Introducing T* and LV-Haystack — our latest leap forward in VLMs for long video understanding! 🧩 Lightweight plugin: T* boosting LLaVA-OV-72B (56→62%) and GPT-4o (50→53%)! ⚡ Fast inference: 34.9s → 10.4s latency, 691 → 170 TFLOPs v.s. SOTA. 📚 Large-scale dataset: 400 hours of videos + 15,000 samples. → TLDR: Long video understanding needs fine-grained supervision, so we start with exploring "key frames" as the first step 🪜. 1/

🚀 Introducing T* and LV-Haystack — our latest leap forward in VLMs for long video understanding! 🧩 Lightweight plugin: T* boosting LLaVA-OV-72B (56→62%) and GPT-4o (50→53%)! ⚡ Fast inference: 34.9s → 10.4s latency, 691 → 170 TFLOPs v.s. SOTA. 📚 Large-scale dataset: 400 hours of videos + 15,000 samples. → TLDR: Long video understanding needs fine-grained supervision, so we start with exploring "key frames" as the first step 🪜. 1/

Zihan "Zenus" Wang

49,615 次观看 • 1 年前

Introducing Einstein Service Agent on Salesforce's new Agentforce Platform! Experience intelligent customer support with seamless collaboration between digital & human workers for rapid case resolution. 🔹 Seamless transfer & human handoffs across all channels 🔹 No writing conversational trees 🔹 Reads text, images, audio/video 🔹 Omni-channel transfer & human handoffs 🔹 Acts across sales, service, marketing, & commerce 🔹 Hybrid keyword & vector search 🔹 Extensible low-code tools 🔹 MuleSoft for cross-system actions 🔹 Real-time, harmonized data with Data Cloud See it at Dreamforce 2024! Now everyone is an Einstein. ❤️

Introducing Einstein Service Agent on Salesforce's new Agentforce Platform! Experience intelligent customer support with seamless collaboration between digital & human workers for rapid case resolution. 🔹 Seamless transfer & human handoffs across all channels 🔹 No writing conversational trees 🔹 Reads text, images, audio/video 🔹 Omni-channel transfer & human handoffs 🔹 Acts across sales, service, marketing, & commerce 🔹 Hybrid keyword & vector search 🔹 Extensible low-code tools 🔹 MuleSoft for cross-system actions 🔹 Real-time, harmonized data with Data Cloud See it at Dreamforce 2024! Now everyone is an Einstein. ❤️

Marc Benioff

358,625 次观看 • 2 年前

🐁🐘🐿🔥SuperAnimals for #Pose -No human labeling - Video analysis on over 45 species with only 2 global classes of animal pose models -If SA needs fine-tuning, its 10× more data efficient & 2X outperforms prior transfer learning 📝Shaokai Ye et al 🧵⬇️

🐁🐘🐿🔥SuperAnimals for #Pose -No human labeling - Video analysis on over 45 species with only 2 global classes of animal pose models -If SA needs fine-tuning, its 10× more data efficient & 2X outperforms prior transfer learning 📝Shaokai Ye et al 🧵⬇️

Mackenzie Weygandt Mathis, PhD

81,656 次观看 • 3 年前

"Girl slams Bageshwar Baba." 🔥🤡🚩 In a video, Bageshwar Baba justified Asaram, who is a convicted rapist of a minor girl and a woman, along with other serious offences. How can a sane person justify a monster and paedophile like Asaram? Shameful!

"Girl slams Bageshwar Baba." 🔥🤡🚩 In a video, Bageshwar Baba justified Asaram, who is a convicted rapist of a minor girl and a woman, along with other serious offences. How can a sane person justify a monster and paedophile like Asaram? Shameful!

Suraj Kumar Bauddh

31,077 次观看 • 5 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

178,382 次观看 • 1 年前

Introducing Prompt-to-leaderboard (P2L): a real-time LLM leaderboard tailored exactly to your use case! P2L trains an LLM to generate "prompt-specific" leaderboards, so you can input a prompt and get a leaderboard specifically for that prompt. The model is trained on the 2M human preference votes from Chatbot Arena. P2L Highlights: 🔹Instant leaderboard for any prompt 🗿 🔹Optimal model routing (hit #1 on Chatbot Arena in Jan 2025 with 1395 score 🧏) 🔹Fine-grained model strength & weakness analysis 🤓 Check out our demo and thread below for more details!

Introducing Prompt-to-leaderboard (P2L): a real-time LLM leaderboard tailored exactly to your use case! P2L trains an LLM to generate "prompt-specific" leaderboards, so you can input a prompt and get a leaderboard specifically for that prompt. The model is trained on the 2M human preference votes from Chatbot Arena. P2L Highlights: 🔹Instant leaderboard for any prompt 🗿 🔹Optimal model routing (hit #1 on Chatbot Arena in Jan 2025 with 1395 score 🧏) 🔹Fine-grained model strength & weakness analysis 🤓 Check out our demo and thread below for more details!

Arena.ai

122,525 次观看 • 1 年前