Loading video...

Video Failed to Load

Go Home

🧵1/2 Cam movement reconstructs the env using photometric alignment features, excluding noisy pixels. Objects (bikes) are reconstructed multiple times because the odometry algorithm doesn't recognize repeated locations, with min error accumulation over time 👇Paper + Code

tetsuoai's profile picture

tetsuo

237,244 subscribers

10,262 views • 2 years ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

We trained a foundation model on 18 million heart ultrasound videos to predict structure instead of pixels. Introducing EchoJEPA, the first foundation-scale JEPA for medical video. Paper: Code: 🧵 1/n
0:40

We trained a foundation model on 18 million heart ultrasound videos to predict structure instead of pixels. Introducing EchoJEPA, the first foundation-scale JEPA for medical video. Paper: Code: 🧵 1/n

Alif Munim (d/acc)

591,427 views • 8 months ago

I don't post dealer positioning often. $SPY $QQQ Not because it doesn't matter. Because most of the time it can read noisy if you don't pair it with other data. But right now it looks like the greeks are showing something worth our attention. Let's take a look together. 🧵
0:20

I don't post dealer positioning often. $SPY $QQQ Not because it doesn't matter. Because most of the time it can read noisy if you don't pair it with other data. But right now it looks like the greeks are showing something worth our attention. Let's take a look together. 🧵

Deep Inference

74,378 views • 2 months ago

Hot take: robots should not dream in pixels. Pixels are too low-level. Latents are too opaque. μ₀ predicts a third thing: 3D motion traces. On real robots, it beats π₀.₅ — with ~1/100 the data scale and no action labels for world-model pretraining. 🧵 (this video features voiceover narration)
2:48

Hot take: robots should not dream in pixels. Pixels are too low-level. Latents are too opaque. μ₀ predicts a third thing: 3D motion traces. On real robots, it beats π₀.₅ — with ~1/100 the data scale and no action labels for world-model pretraining. 🧵 (this video features voiceover narration)

Furong Huang

123,855 views • 4 months ago

It's time to recognize the risks, protect young players, and make heading safer. Repeated head impacts can have lasting consequences for brain health. Just because we can't see the danger doesn't mean it isn't there.
1:19

It's time to recognize the risks, protect young players, and make heading safer. Repeated head impacts can have lasting consequences for brain health. Just because we can't see the danger doesn't mean it isn't there.

Raphaël Varane

159,006 views • 3 months ago

1/7 I'm super excited to share LidarVerse! One model that reconstructs, forecasts and generates LiDAR across sensors, in real time. 📄 Paper: 🌐 Project: 💻 Code: 🤗 Models:
0:32

1/7 I'm super excited to share LidarVerse! One model that reconstructs, forecasts and generates LiDAR across sensors, in real time. 📄 Paper: 🌐 Project: 💻 Code: 🤗 Models:

Shengyu Huang

16,271 views • 10 days ago

The video highlights the one-inch punch, showing how precise technique and body alignment create power at close range, with a master emphasizing guidance over repetition using a simple paper drill.
0:43

The video highlights the one-inch punch, showing how precise technique and body alignment create power at close range, with a master emphasizing guidance over repetition using a simple paper drill.

Massimo

38,180 views • 9 months ago

[SIGGRAPH '24] RTG-SLAM: Real-time 3D Reconstruction at Scale Using Gaussian Splatting Paper (pdf link): Project: Code: Method ⬇️ 1 I 2
1:00

[SIGGRAPH '24] RTG-SLAM: Real-time 3D Reconstruction at Scale Using Gaussian Splatting Paper (pdf link): Project: Code: Method ⬇️ 1 I 2

MrNeRF

35,735 views • 2 years ago

a version 1 QR code with maximum error correction can store 7 bytes. That’s 2⁵⁶ possible codes. Here are the first 1,043 of them-
0:55

a version 1 QR code with maximum error correction can store 7 bytes. That’s 2⁵⁶ possible codes. Here are the first 1,043 of them-

Matt Henderson

175,433 views • 8 months ago

Introducing Objaverse-XL, an open dataset of over 10 million 3D objects! With it, we train Zero123-XL, a foundation model for 3D, observing incredible 3D generalization abilities: 🧵👇 📝 Paper:
0:53

Introducing Objaverse-XL, an open dataset of over 10 million 3D objects! With it, we train Zero123-XL, a foundation model for 3D, observing incredible 3D generalization abilities: 🧵👇 📝 Paper:

Matt Deitke

368,910 views • 3 years ago

Introducing Objects, the last step to your workflow 🔥 Add the same item across multiple images, and combine it with the rest of the tools in Freepik AI Suite Step-by-step tutorial below 🧵👇
0:53

Introducing Objects, the last step to your workflow 🔥 Add the same item across multiple images, and combine it with the rest of the tools in Freepik AI Suite Step-by-step tutorial below 🧵👇

Freepik

53,759 views • 1 year ago

Many Are Still Either In Denial Or Still Asleep To Not Recognize All The Obvious Signs We Are In The End Times. It’s Time To Be Ready! 👇
8:03

Many Are Still Either In Denial Or Still Asleep To Not Recognize All The Obvious Signs We Are In The End Times. It’s Time To Be Ready! 👇

PaulsCorner-VerseQuest

56,042 views • 1 year ago

More than a few dozen teens on bikes took over a mall just a few miles from Los Angeles. It appears this isn’t the first time. Similar takeovers have happened there multiple times before.
0:52

More than a few dozen teens on bikes took over a mall just a few miles from Los Angeles. It appears this isn’t the first time. Similar takeovers have happened there multiple times before.

I Meme Therefore I Am 🇺🇸

31,442 views • 4 months ago

Birthday Story Prequel (3 min 1 sec) Made using Zeroscope and Bark. Few tech notes in the 🧵
3:01

Birthday Story Prequel (3 min 1 sec) Made using Zeroscope and Bark. Few tech notes in the 🧵

Roope Rainisto

36,029 views • 3 years ago

In the western world, people would not recognize a lot of what goes on in the third world cultures. Bringing the first world to the third often times doesn't work because the people have no grasp of why things are the way they are.
1:30

In the western world, people would not recognize a lot of what goes on in the third world cultures. Bringing the first world to the third often times doesn't work because the people have no grasp of why things are the way they are.

Ian Miles

1,788,211 views • 1 year ago

🦾 Our robot hand can rotate objects over 6+ axes in the real-world! Introducing RotateIt (CoRL 2023), a Sim-to-Real policy that can rotate many objects over many axes, using vision and touch! Check it out: Paper: #CoRL2023
0:25

🦾 Our robot hand can rotate objects over 6+ axes in the real-world! Introducing RotateIt (CoRL 2023), a Sim-to-Real policy that can rotate many objects over many axes, using vision and touch! Check it out: Paper: #CoRL2023

Haozhi Qi

51,041 views • 3 years ago

GM web3! Finished my last Hyperfy world of the year. Used many of the amazing features to create a little VR adventure game with many avatars, vehicles and objects to interact with. I'm so excited for 2.0 version coming soon! Play or just explore. There are prizes👇🧵
0:53

GM web3! Finished my last Hyperfy world of the year. Used many of the amazing features to create a little VR adventure game with many avatars, vehicles and objects to interact with. I'm so excited for 2.0 version coming soon! Play or just explore. There are prizes👇🧵

Amargo

55,579 views • 1 year ago

🧵1/2 MIT's Charles Leiserson covers the progression from source code to compilation, machine code, hardware interpretation, and ultimately, execution. 👇Full MIT Performance Engineering of Software Systems Course 👇
1:16:33

🧵1/2 MIT's Charles Leiserson covers the progression from source code to compilation, machine code, hardware interpretation, and ultimately, execution. 👇Full MIT Performance Engineering of Software Systems Course 👇

tetsuo

115,218 views • 2 years ago

Introducing GRID: the General Robot Intelligence Development platform, designed for prototyping smart and safe robots rapidly using foundation models, LLMs, and simulation. Paper: Try now: GitHub: 🧵👇(1/N)
0:22

Introducing GRID: the General Robot Intelligence Development platform, designed for prototyping smart and safe robots rapidly using foundation models, LLMs, and simulation. Paper: Try now: GitHub: 🧵👇(1/N)

Sai Vemprala

277,340 views • 3 years ago