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。

Giving Microduck object-tracking vision capabilities! 🦆👀🤖 It can now play fetch and doesn't get confused by other objects. Using Roboflow Trackers library I made Microduck keep track of the different objects in its sight and has a control policy for reaching the ball it tracks. Open-Source code below 👇

Alex Bodner

1,604 subscribers

31,551 次观看 • 20 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

11 条评论

Alex Bodner 的头像

Alex Bodner20 天前

object tracking library: microduck integration: Leave a star🌟 if you like it!

jrubiosainz 的头像

jrubiosainz19 天前

@roboflow Oh!! This is quite similar concept than rhe one I created here:

Alex Bodner 的头像

Alex Bodner19 天前

@roboflow Nice!

Konstantin Goncharuk 🇺🇦 的头像

Konstantin Goncharuk 🇺🇦20 天前

@roboflow looks very cool! is there also an example how to use it with the sim you have on the video? whats that btw?

Alex Bodner 的头像

Alex Bodner20 天前

@roboflow thanks! yes, you can replicate the video with:

Konstantin Goncharuk 🇺🇦 的头像

Konstantin Goncharuk 🇺🇦20 天前

@roboflow ah, awesome, thanks!

Moira Shooter 的头像

Moira Shooter20 天前

@roboflow This is really cool! 😎

Alex Bodner 的头像

Alex Bodner20 天前

@roboflow Thank you!

Rok Benko 的头像

Rok Benko19 天前

I open-sourced quackd: a brain for the Microduck. You tell the Microduck what you want in plain language, like "find the ball and kick it", and an LLM plans the steps using the skills the duck already has. Supports local LLMs as well. Goals live in .duck files, the same idea as the .skill files Claude uses, but for a duck. Works today in the bundled simulator, no hardware needed.

James Astor 的头像

James Astor19 天前

@matth_lapeyre @roboflow How many policies can the microduck have in the end? Like, once it actually arrives will it be capable of tens or hundreds of capabilities?

SB 的头像

SB18 天前

@roboflow Inspired by your elegant two-view setup—we added a true head-camera POV alongside an elevated overview for our controlled-roll experiment. The 0.82 s handoff passed 3/3 perturbed seeds. Open results:

相关视频

To me, microduck looks like a droid. So I made it sound like one Microduck is open source and hackable. That means you can program it to do whatever you want! I trained a 1.3m parameter LLM to output droid like responses. It is running in the browser The model has learned to respond with an emotion, alongside English like Happy, Excited, Sad Then the emotion is fed into a Rust based synthesizer to generate R2D2 like sounds that are aligned with the emotion --- completely procedurally, no wav files, no copyright violation You can chat with it. You can also customize it, give it a name and set your own colors! All set in a spaceship hangar backdrop with other friendly ducks---simulated with realistic physics and microduck's actual walking and skating policies! All in the browser, with WASM!!! I can put literally the SAME code in the browser, and then compile it into an ARM binary and put it in the actual robot in real life 🤯 If you like this and want to see more of it, please like and share so that my boss puts me in the priority list for the first microduck batch 😎 All open source of course! Microquack! 🦆 Hugging Face Space:

To me, microduck looks like a droid. So I made it sound like one Microduck is open source and hackable. That means you can program it to do whatever you want! I trained a 1.3m parameter LLM to output droid like responses. It is running in the browser The model has learned to respond with an emotion, alongside English like Happy, Excited, Sad Then the emotion is fed into a Rust based synthesizer to generate R2D2 like sounds that are aligned with the emotion --- completely procedurally, no wav files, no copyright violation You can chat with it. You can also customize it, give it a name and set your own colors! All set in a spaceship hangar backdrop with other friendly ducks---simulated with realistic physics and microduck's actual walking and skating policies! All in the browser, with WASM!!! I can put literally the SAME code in the browser, and then compile it into an ARM binary and put it in the actual robot in real life 🤯 If you like this and want to see more of it, please like and share so that my boss puts me in the priority list for the first microduck batch 😎 All open source of course! Microquack! 🦆 Hugging Face Space:

Onur Solmaz

33,500 次观看 • 23 天前

Resizing objects in Blender can get surprisingly tedious So, inspired by Paralives and using the amazing assets provided by Kenney, I built a Dynamic Scaling System that handles it automatically! Just adjust the scale and the object adapts itself Available on our Patreon! ✨

Resizing objects in Blender can get surprisingly tedious So, inspired by Paralives and using the amazing assets provided by Kenney, I built a Dynamic Scaling System that handles it automatically! Just adjust the scale and the object adapts itself Available on our Patreon! ✨

DX

78,334 次观看 • 1 个月前

New in Notch 2026.2: fully interactive Gaussian Splats. Imported objects and environments can be lit, reflected, and cast shadows, sitting naturally alongside any other element in the production. In this sample, we used a splat of a library cafe with a few 3D objects dropped inside to show it in action. Grab the sample scene from the Launcher and try it for yourself. Splat by:

New in Notch 2026.2: fully interactive Gaussian Splats. Imported objects and environments can be lit, reflected, and cast shadows, sitting naturally alongside any other element in the production. In this sample, we used a splat of a library cafe with a few 3D objects dropped inside to show it in action. Grab the sample scene from the Launcher and try it for yourself. Splat by:

Notch

11,312 次观看 • 2 个月前

RedNova: From Arbitrary File Write to Command Execution Using COM in Windows Months ago, I wrote a post ( about the unique method used in the RedSun exploit to turn a file write into command execution by using a COM object that can be triggered by a low-privileged user and runs as SYSTEM. In that post, I mentioned that there are other COM objects that can be used in the same way. I presented one object called CoFilterPipeline and noted that it needs some tricks before it can be weaponized. Now I've written a full blog post on how you can find such objects and how you can weaponize them in exploits like RedSun. The video below shows another COM object with the same behavior (windows 11). It's called PerceptionSimulationCoClass. A low-privileged user can activate it, and it runs as SYSTEM. The video doesn't contain any exploit (it just replaces the COM server as administrator). The goal is only to demonstrate how a file write can be turned into command execution as SYSTEM while being triggered by a low-privileged user.

RedNova: From Arbitrary File Write to Command Execution Using COM in Windows Months ago, I wrote a post ( about the unique method used in the RedSun exploit to turn a file write into command execution by using a COM object that can be triggered by a low-privileged user and runs as SYSTEM. In that post, I mentioned that there are other COM objects that can be used in the same way. I presented one object called CoFilterPipeline and noted that it needs some tricks before it can be weaponized. Now I've written a full blog post on how you can find such objects and how you can weaponize them in exploits like RedSun. The video below shows another COM object with the same behavior (windows 11). It's called PerceptionSimulationCoClass. A low-privileged user can activate it, and it runs as SYSTEM. The video doesn't contain any exploit (it just replaces the COM server as administrator). The goal is only to demonstrate how a file write can be turned into command execution as SYSTEM while being triggered by a low-privileged user.

Haidar

13,648 次观看 • 29 天前

everyone is talking about the new kimi k2.7 code so i tested it for real. i built a fruit ninja game i play with my index finger. webcam tracks my finger, turns it into a blade, i swipe and slice fruit midair. real tracking in the browser. and yes it's open source

everyone is talking about the new kimi k2.7 code so i tested it for real. i built a fruit ninja game i play with my index finger. webcam tracks my finger, turns it into a blade, i swipe and slice fruit midair. real tracking in the browser. and yes it's open source

Farhan

29,728 次观看 • 3 个月前

LINGO-1 doesn't just describe what it sees; it can also show us what it sees! 👀 Introducing referential segmentation - a new feature that grounds what #LINGO1 describes to specific objects in the image 🖼💬, allowing for more accurate and trustworthy responses. Watch below👇

LINGO-1 doesn't just describe what it sees; it can also show us what it sees! 👀 Introducing referential segmentation - a new feature that grounds what #LINGO1 describes to specific objects in the image 🖼💬, allowing for more accurate and trustworthy responses. Watch below👇

Wayve

16,425 次观看 • 2 年前

Researchers at Duke University developed Argus, a spherical robot with 20 legs and 20 eye-like sensors that can see and move in any direction. Using a concept called dynamic symmetry, it stays balanced and agile while traveling across rough terrain and climbing obstacles. The robot can also track objects and keep working even if some of its legs or sensors fail, making it useful for exploring dangerous environments and other planets.

Researchers at Duke University developed Argus, a spherical robot with 20 legs and 20 eye-like sensors that can see and move in any direction. Using a concept called dynamic symmetry, it stays balanced and agile while traveling across rough terrain and climbing obstacles. The robot can also track objects and keep working even if some of its legs or sensors fail, making it useful for exploring dangerous environments and other planets.

Space and Technology

23,452 次观看 • 3 个月前

GM Microduck Family! 🦆 The Microduck Custom PFP Event has officially come to an end! First of all, thank you so much to everyone who participated. Seeing so many people customize their own Microduck PFPs, create memes, and have fun with the character was genuinely amazing. This event was all about bringing the Microduck community together and seeing everyone put their own personality into Microduck — and you guys absolutely delivered. ❤️ I also want to give a special thank you to chit🇰🇷 for helping me organize and run this event. You helped bring the Microduck community together, and I truly appreciate everything you did. 🦆 Now, here are the winners: 🏆 GRAND PRIZE — $1,200 Roar Web3man Originally, every winner was going to receive $400. However, Roar Web3man put an incredible amount of effort into the entry and even took Microduck into the real world to promote it on the streets. We loved the dedication so much that we decided to award 3x the original prize — $1,200. 🦆🔥 🎨 WINNERS — $400 EACH @todayskill_ai Degenzo 👑 JIN Romeo_sNp Yaho Jones__ Mandu @Jinchwa Humbleman chris_Han 📢 TWO WINNERS WE HAVEN'T BEEN ABLE TO REACH The following two entries were also selected as winners, but I haven't been able to get in touch with them yet: dododo dpi If either of you sees this post, please leave a reply under the entry I selected. If I don't hear from you within 5 hours of this post, I will select new winners in your place. Finally, a huge thank you again to everyone who helped make this event possible: Pika XBT Miyamoto MoneyLord Kaduna 🦅 Eric Cryptoman And of course, thank you to every single person who participated. The creativity, memes, and energy you brought to this event were incredible. I hope we can keep building and having fun together as the Microduck community grows. 🦆❤️ And once again — congratulations to all the winners! 🦆🏆

GM Microduck Family! 🦆 The Microduck Custom PFP Event has officially come to an end! First of all, thank you so much to everyone who participated. Seeing so many people customize their own Microduck PFPs, create memes, and have fun with the character was genuinely amazing. This event was all about bringing the Microduck community together and seeing everyone put their own personality into Microduck — and you guys absolutely delivered. ❤️ I also want to give a special thank you to chit🇰🇷 for helping me organize and run this event. You helped bring the Microduck community together, and I truly appreciate everything you did. 🦆 Now, here are the winners: 🏆 GRAND PRIZE — $1,200 Roar Web3man Originally, every winner was going to receive $400. However, Roar Web3man put an incredible amount of effort into the entry and even took Microduck into the real world to promote it on the streets. We loved the dedication so much that we decided to award 3x the original prize — $1,200. 🦆🔥 🎨 WINNERS — $400 EACH @todayskill_ai Degenzo 👑 JIN Romeo_sNp Yaho Jones__ Mandu @Jinchwa Humbleman chris_Han 📢 TWO WINNERS WE HAVEN'T BEEN ABLE TO REACH The following two entries were also selected as winners, but I haven't been able to get in touch with them yet: dododo dpi If either of you sees this post, please leave a reply under the entry I selected. If I don't hear from you within 5 hours of this post, I will select new winners in your place. Finally, a huge thank you again to everyone who helped make this event possible: Pika XBT Miyamoto MoneyLord Kaduna 🦅 Eric Cryptoman And of course, thank you to every single person who participated. The creativity, memes, and energy you brought to this event were incredible. I hope we can keep building and having fun together as the Microduck community grows. 🦆❤️ And once again — congratulations to all the winners! 🦆🏆

MicroDuck

19,959 次观看 • 17 天前

🤖 Ever seen a robot get confused by a glass door? That era is ending. Today, we're open-sourcing LingBot-Depth, a new spatial perception model that lets robots truly *see* the physical world, including transparent & reflective objects. It even surpasses top-tier industrial depth cameras in core metrics. A new sense of sight for embodied AI. #EmbodiedAI #Robotics #ComputerVision #OpenSource #DepthCamera

🤖 Ever seen a robot get confused by a glass door? That era is ending. Today, we're open-sourcing LingBot-Depth, a new spatial perception model that lets robots truly see the physical world, including transparent & reflective objects. It even surpasses top-tier industrial depth cameras in core metrics. A new sense of sight for embodied AI. #EmbodiedAI #Robotics #ComputerVision #OpenSource #DepthCamera

Robbyant

320,159 次观看 • 7 个月前

Grok Imagine gives you complete control over your videos, letting you direct the camera's movement, control your character's actions, script their dialogue, and manage how objects in the scene interact. Your characters can smash objects, swing a bat, or walk out of a room. To build seamless videos, start by extracting the last frame from your current clip and using it as the starting frame for the next one.

Grok Imagine gives you complete control over your videos, letting you direct the camera's movement, control your character's actions, script their dialogue, and manage how objects in the scene interact. Your characters can smash objects, swing a bat, or walk out of a room. To build seamless videos, start by extracting the last frame from your current clip and using it as the starting frame for the next one.

tetsuo

27,443 次观看 • 11 个月前

I DON’T THINK PEOPLE REALIZE HOW MASSIVE THIS IS YET Hugging Face and Pollen Robotics just released Microduck. It is a new legged robot for $399. It passed $1 million in sales in under SEVEN hours 🤯 The hardware packs 15 actuators, a camera, and LiDAR into a 25 cm frame. It weighs under 800 grams and runs for about an hour on a single charge. Out of the box, it can > walk > roller-skate > pick up objects > recover from falls .. and so much more! You can drive the 7 shipped policies from the first boot using a gamepad. But the real hook is the training stack. The software is entirely open: → MuJoCo simulation and PPO training → 50 Hz onboard policy loop → Browser simulator included alongside hardware → Full control stack available on GitHub The mechanical and electronic design files are still proprietary. But don’t let the cute robotic duck fool you. Under the closed chassis is a FULLY OPEN-SOURCE, hackable brain you can experiment with and build on. Repo in 🧵↓

I DON’T THINK PEOPLE REALIZE HOW MASSIVE THIS IS YET Hugging Face and Pollen Robotics just released Microduck. It is a new legged robot for $399. It passed $1 million in sales in under SEVEN hours 🤯 The hardware packs 15 actuators, a camera, and LiDAR into a 25 cm frame. It weighs under 800 grams and runs for about an hour on a single charge. Out of the box, it can > walk > roller-skate > pick up objects > recover from falls .. and so much more! You can drive the 7 shipped policies from the first boot using a gamepad. But the real hook is the training stack. The software is entirely open: → MuJoCo simulation and PPO training → 50 Hz onboard policy loop → Browser simulator included alongside hardware → Full control stack available on GitHub The mechanical and electronic design files are still proprietary. But don’t let the cute robotic duck fool you. Under the closed chassis is a FULLY OPEN-SOURCE, hackable brain you can experiment with and build on. Repo in 🧵↓

Charly Wargnier ♨️

651,785 次观看 • 22 天前

💡 I am building an Apple Vision Pro app to control my home lights — and it remembers where I placed the controls, even after rebooting. Using ARKit’s World Anchors in a full space, the app persists virtual objects across launches and reboots. Now I just look at a light and toggle it on/off. Set it up once. Feels like the controls are part of my space. Thinking of making a tutorial — would that be helpful? 👇

💡 I am building an Apple Vision Pro app to control my home lights — and it remembers where I placed the controls, even after rebooting. Using ARKit’s World Anchors in a full space, the app persists virtual objects across launches and reboots. Now I just look at a light and toggle it on/off. Set it up once. Feels like the controls are part of my space. Thinking of making a tutorial — would that be helpful? 👇

Sarang Borude

86,473 次观看 • 1 年前

We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny open-source biped with 15 actuators and packed with sensors (camera, speaker, LiDAR, NFC, bluetooth, wifi, etc) that you train yourself with reinforcement learning. It's also playable out of the box with more than half a dozen fun and playful pre-trained policies to have it walk, sit, crouch, roller-skate, pick up objects with its articulated beak, and recover on its own. And all for less than $400. See all the details, play with the simulator and order it at: (video with sound on 🔊)

We have a huge news to share today! Today we are unveiling the first truly accessible RL robot - welcome Microduck A 25 cm tiny open-source biped with 15 actuators and packed with sensors (camera, speaker, LiDAR, NFC, bluetooth, wifi, etc) that you train yourself with reinforcement learning. It's also playable out of the box with more than half a dozen fun and playful pre-trained policies to have it walk, sit, crouch, roller-skate, pick up objects with its articulated beak, and recover on its own. And all for less than $400. See all the details, play with the simulator and order it at: (video with sound on 🔊)

Thomas Wolf

2,887,041 次观看 • 25 天前

It’s incredible that you can let Opus 5 run for two hours in Claude Code and get a cinematic Three.js movie inside one HTML file. It created the story, scenes, transitions, camera moves, and procedural world. Higgsfield generated the narration, sound effects, and music. I mostly helped with references for more realistic grass, added the skills I use, and asked it to keep verifying its work. It’s all open source, so you can play with the code. HTML Site: GitHub: Grass reference:

It’s incredible that you can let Opus 5 run for two hours in Claude Code and get a cinematic Three.js movie inside one HTML file. It created the story, scenes, transitions, camera moves, and procedural world. Higgsfield generated the narration, sound effects, and music. I mostly helped with references for more realistic grass, added the skills I use, and asked it to keep verifying its work. It’s all open source, so you can play with the code. HTML Site: GitHub: Grass reference:

Meng To

68,404 次观看 • 1 个月前

BREAKING🚨: A mysterious object appears near the Moon’s surface, creating a moment that looks unusual and difficult to explain at first. The bright lunar background makes even small moving objects clearly visible. It may look like the object is flying very close to the Moon, almost touching its surface. But in reality, it is often much closer to Earth, and this is just a result of alignment and perspective. Objects like satellites, meteoroids, or space debris can pass in front of the Moon, appearing as dark moving shapes for a few seconds.

BREAKING🚨: A mysterious object appears near the Moon’s surface, creating a moment that looks unusual and difficult to explain at first. The bright lunar background makes even small moving objects clearly visible. It may look like the object is flying very close to the Moon, almost touching its surface. But in reality, it is often much closer to Earth, and this is just a result of alignment and perspective. Objects like satellites, meteoroids, or space debris can pass in front of the Moon, appearing as dark moving shapes for a few seconds.

The Astronomy Guy

32,892 次观看 • 5 个月前

You can now run Claude Code locally for FREE. → Using open-source models → Locally and fully private → Full agent + tool workflows Here's the FULL step-by-step tutorial to get it all set up in less than 10 minutes:

You can now run Claude Code locally for FREE. → Using open-source models → Locally and fully private → Full agent + tool workflows Here's the FULL step-by-step tutorial to get it all set up in less than 10 minutes:

Alvaro Cintas

1,319,857 次观看 • 8 个月前

Open Source has done it again. AI at Meta have released the code for their new Animated Drawings tool. AI can now automatically animate children's drawings of human-like figures. The demo is free, the research paper is available and the code and dataset (nearly 180k annotated amateur drawings) is public. More below 👇

Open Source has done it again. AI at Meta have released the code for their new Animated Drawings tool. AI can now automatically animate children's drawings of human-like figures. The demo is free, the research paper is available and the code and dataset (nearly 180k annotated amateur drawings) is public. More below 👇

d@x

49,061 次观看 • 3 年前

This should flip and go the other way if you stare at it for a bit! I love open source... a couple of days ago I posted the code to this Dyson Ring that I wrote, but I couldn't quite get the illusion right. Today, Sean Lavery checked in a double-sided rendering and some other changes that really sell it. Code is at based on a demo by Joseff

This should flip and go the other way if you stare at it for a bit! I love open source... a couple of days ago I posted the code to this Dyson Ring that I wrote, but I couldn't quite get the illusion right. Today, Sean Lavery checked in a double-sided rendering and some other changes that really sell it. Code is at based on a demo by Joseff

Dave W Plummer

25,402 次观看 • 9 个月前