Loading video...

Video Failed to Load

Go Home

Share our recent work - TrackingWorld: World-centric Monocular 3D Tracking of Almost All Pixels [NeurIPS'25] An optimization-based 3D tracking of all pixels. Code: Webpage: Welcome to try it! A feedforward version is on the way!

YuanLiu41955461's profile picture

Yuan Liu

2,007 subscribers

12,165 views • 8 months 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

SpatialTracker Tracking Any 2D Pixels in 3D Space Recovering dense and long-range pixel motion in videos is a challenging problem. Part of the difficulty arises from the 3D-to-2D projection process, leading to occlusions and discontinuities in the 2D motion domain.
0:11

SpatialTracker Tracking Any 2D Pixels in 3D Space Recovering dense and long-range pixel motion in videos is a challenging problem. Part of the difficulty arises from the 3D-to-2D projection process, leading to occlusions and discontinuities in the 2D motion domain.

AK

77,162 views • 2 years ago

3d world abilities of AI models keep getting better, this is @Kling_AI 1.5 It can move around a space in 3d and you can steer its camera movements All based on a single input image
0:10

3d world abilities of AI models keep getting better, this is @Kling_AI 1.5 It can move around a space in 3d and you can steer its camera movements All based on a single input image

@levelsio

112,201 views • 1 year ago

Our map tracking Russia's full-scale invasion of Ukraine has almost more than 20,000 geolocated videos and photos since 2022. All of the data is interactive and traceable on the map here: 👇
0:12

Our map tracking Russia's full-scale invasion of Ukraine has almost more than 20,000 geolocated videos and photos since 2022. All of the data is interactive and traceable on the map here: 👇

Benjamin Strick

160,165 views • 2 years ago

hi all, excited to join! i'm building an expressive mini "shoggoth" robot which will eventually be hooked up to gpt4o realtime voice. i'm currently working on the low-level policies, which are trained in a mujoco simulation with RL. to delay working on raw-pixels for now, i trained a pose-estimation model using deeplabcut and triangulate the position in 3d space using the stereo cameras. eventually, i'll use gpt4o's tool calling capabilities to activate several of these policies (closed and open loop) based on the dialog flow! captions: manual actuation of the tentacle / 3d pose estimation / target design
0:29

hi all, excited to join! i'm building an expressive mini "shoggoth" robot which will eventually be hooked up to gpt4o realtime voice. i'm currently working on the low-level policies, which are trained in a mujoco simulation with RL. to delay working on raw-pixels for now, i trained a pose-estimation model using deeplabcut and triangulate the position in 3d space using the stereo cameras. eventually, i'll use gpt4o's tool calling capabilities to activate several of these policies (closed and open loop) based on the dialog flow! captions: manual actuation of the tentacle / 3d pose estimation / target design

Matthieu LC

45,587 views • 1 year ago

Finally... A Tracking system that is similar to Vbridger for 3D vtuber Luwke developed this Vbridger-adjacent system for 3D Vtuber and taught me It shares the same principles of Vbridger, which allow for a more custom look and get even closer to its 2D counterpart #ShylilyArt
0:13

Finally... A Tracking system that is similar to Vbridger for 3D vtuber Luwke developed this Vbridger-adjacent system for 3D Vtuber and taught me It shares the same principles of Vbridger, which allow for a more custom look and get even closer to its 2D counterpart #ShylilyArt

Azor Phasecrown • 3D Modeller • Comm Open

46,419 views • 2 months ago

2025 is year of 3D AI Rodin AI 1.5 now can not only generate 3D model with clean meshes but also create high quality textures You can export all the models to 3D software to create a new world
0:18

2025 is year of 3D AI Rodin AI 1.5 now can not only generate 3D model with clean meshes but also create high quality textures You can export all the models to 3D software to create a new world

el.cine

92,563 views • 1 year ago

Gaussian splats don't simulate light. They capture it. I took one photograph of this room and generated it as a fully walkable 3D world in Mint. The volumetric rays, the shadow play on the rug, all preserved. This is what real-time 3D is becoming
0:13

Gaussian splats don't simulate light. They capture it. I took one photograph of this room and generated it as a fully walkable 3D world in Mint. The volumetric rays, the shadow play on the rug, all preserved. This is what real-time 3D is becoming

mint

15,446 views • 2 months ago

oh my.. this is over for developers Gemini 3 can create interactive 3D webpage in mins, just a few simple text prompts, it generates all the code you can control millions of particles with your hands and make them form any shape you want tutorial and prompts below:
0:18

oh my.. this is over for developers Gemini 3 can create interactive 3D webpage in mins, just a few simple text prompts, it generates all the code you can control millions of particles with your hands and make them form any shape you want tutorial and prompts below:

el.cine

616,735 views • 8 months ago

I've been working on a custom tool to visualize our compiler optimization graphs for SpiderMonkey. The layout algorithm is all custom and produces dramatically more readable and stable graphs than GraphViz. All the layout and rendering is less than 1000 lines of code.
0:17

I've been working on a custom tool to visualize our compiler optimization graphs for SpiderMonkey. The layout algorithm is all custom and produces dramatically more readable and stable graphs than GraphViz. All the layout and rendering is less than 1000 lines of code.

Ben Visness

23,899 views • 1 year ago

Static 3D generation isn't enough. We need assets ready for animation. Our new #SIGGRAPH work, AniGen, takes a single image and generates the 3D shape, skeleton, and skinning weights all at once. Code is fully open-sourced! Kudos to Yihua and VAST AI Research 🧵(1/4)
0:15

Static 3D generation isn't enough. We need assets ready for animation. Our new #SIGGRAPH work, AniGen, takes a single image and generates the 3D shape, skeleton, and skinning weights all at once. Code is fully open-sourced! Kudos to Yihua and VAST AI Research 🧵(1/4)

Yanpei Cao

145,188 views • 4 months ago

🚀🌍🚀 Echo-2 is live! It generates high-quality physically grounded 3D scenes instead of just pixels. Check out the open demo to create and explore worlds in real-time! I gave it a picture from my Himalayas trek—it handles the massive scale range really well! 🏔️🌍
0:12

🚀🌍🚀 Echo-2 is live! It generates high-quality physically grounded 3D scenes instead of just pixels. Check out the open demo to create and explore worlds in real-time! I gave it a picture from my Himalayas trek—it handles the massive scale range really well! 🏔️🌍

David Novotny

11,435 views • 4 months ago

Charity powered by Web3SlashPlace People have lost everything and are suffering from the horrible floods in Northern Thailand. 🆘 Andrew and his girlfriend went to the temple to donate food, water, clothes, etc. Once enough funds are raised Andrew plans on going directly to the northern Thai/Myanmar border to help the people suffering from the civil war in Myanmar and others suffering from the floods as well. 💚 During this evenings space we will have a discount on pixels for Web3SlashPlace where any amount of pixels purchased in SOL, ETH, and/or MATIC will be doubled! So for example, if you buy 100 pixels we will airdrop an additional 100 pixels to you manually after the space.▫️ 20% of all funds raised through Web3SlashPlace will go directly to people who need it most. We have ensured the donations today are going to those in need directly through a friend of ours who is doing more hands on work to help the tens of thousands of people (or more) effected by these floods. 🇹🇭
0:12

Charity powered by Web3SlashPlace People have lost everything and are suffering from the horrible floods in Northern Thailand. 🆘 Andrew and his girlfriend went to the temple to donate food, water, clothes, etc. Once enough funds are raised Andrew plans on going directly to the northern Thai/Myanmar border to help the people suffering from the civil war in Myanmar and others suffering from the floods as well. 💚 During this evenings space we will have a discount on pixels for Web3SlashPlace where any amount of pixels purchased in SOL, ETH, and/or MATIC will be doubled! So for example, if you buy 100 pixels we will airdrop an additional 100 pixels to you manually after the space.▫️ 20% of all funds raised through Web3SlashPlace will go directly to people who need it most. We have ensured the donations today are going to those in need directly through a friend of ours who is doing more hands on work to help the tens of thousands of people (or more) effected by these floods. 🇹🇭

Zack Morris

13,267 views • 1 year ago

I'll be opening 3D Vtuber commissions on the 30th of this month (bear in mind this could be subject to change) in the meantime I'll work on redoing some renders for my 3D vtuber portfolio. Here is a vid to give you an idea of my work. If interested feel free to dm me! 👍
0:10

I'll be opening 3D Vtuber commissions on the 30th of this month (bear in mind this could be subject to change) in the meantime I'll work on redoing some renders for my 3D vtuber portfolio. Here is a vid to give you an idea of my work. If interested feel free to dm me! 👍

Mr Zing

27,330 views • 11 months ago

Meta releases VGGSfM Visual Geometry Grounded Deep Structure From Motion Structure-from-motion (SfM) is a long-standing problem in the computer vision community, which aims to reconstruct the camera poses and 3D structure of a scene from a set of unconstrained 2D images. Classical frameworks solve this problem in an incremental manner by detecting and matching keypoints, registering images, triangulating 3D points, and conducting bundle adjustment. Recent research efforts have predominantly revolved around harnessing the power of deep learning techniques to enhance specific elements (e.g., keypoint matching), but are still based on the original, non-differentiable pipeline. Instead, we propose a new deep SfM pipeline VGGSfM, where each component is fully differentiable and thus can be trained in an end-to-end manner. To this end, we introduce new mechanisms and simplifications. First, we build on recent advances in deep 2D point tracking to extract reliable pixel-accurate tracks, which eliminates the need for chaining pairwise matches. Furthermore, we recover all cameras simultaneously based on the image and track features instead of gradually registering cameras. Finally, we optimise the cameras and triangulate 3D points via a differentiable bundle adjustment layer. We attain state-of-the-art performance on three popular datasets, CO3D, IMC Phototourism, and ETH3D.
0:13

Meta releases VGGSfM Visual Geometry Grounded Deep Structure From Motion Structure-from-motion (SfM) is a long-standing problem in the computer vision community, which aims to reconstruct the camera poses and 3D structure of a scene from a set of unconstrained 2D images. Classical frameworks solve this problem in an incremental manner by detecting and matching keypoints, registering images, triangulating 3D points, and conducting bundle adjustment. Recent research efforts have predominantly revolved around harnessing the power of deep learning techniques to enhance specific elements (e.g., keypoint matching), but are still based on the original, non-differentiable pipeline. Instead, we propose a new deep SfM pipeline VGGSfM, where each component is fully differentiable and thus can be trained in an end-to-end manner. To this end, we introduce new mechanisms and simplifications. First, we build on recent advances in deep 2D point tracking to extract reliable pixel-accurate tracks, which eliminates the need for chaining pairwise matches. Furthermore, we recover all cameras simultaneously based on the image and track features instead of gradually registering cameras. Finally, we optimise the cameras and triangulate 3D points via a differentiable bundle adjustment layer. We attain state-of-the-art performance on three popular datasets, CO3D, IMC Phototourism, and ETH3D.

AK

96,527 views • 2 years ago

This isn’t a 3D model thrown in a box. This is pure code. No Blender, no Maya, no shortcuts. Every voxel, every jump, every movement scripted from scratch. Generative art at this level means thinking like a designer, a coder, and an artist all at once. It’s not about placing pixels. It’s about writing the rules of how a frog should exist in the Bitcoin chain. BLOCKFROGS isn’t just generative. It’s engineered. Be part of this experience Wl = 2 Freemints 5 WL to snatch Post or Quote about BLOCK | Frogs | Panthers Drop the link
0:18

This isn’t a 3D model thrown in a box. This is pure code. No Blender, no Maya, no shortcuts. Every voxel, every jump, every movement scripted from scratch. Generative art at this level means thinking like a designer, a coder, and an artist all at once. It’s not about placing pixels. It’s about writing the rules of how a frog should exist in the Bitcoin chain. BLOCKFROGS isn’t just generative. It’s engineered. Be part of this experience Wl = 2 Freemints 5 WL to snatch Post or Quote about BLOCK | Frogs | Panthers Drop the link

BLOCK | Frogs | Panthers

16,246 views • 1 year ago

Code LLaMA is now on Perplexity’s LLaMa Chat! Try asking it to write a function for you, or explain a code snippet: 🔗 This is the fastest way to try AI at Meta’s latest code-specialized LLM. With our model deployment expertise, we are able to provide you with this model less than 24 hours of it’s release. What’s next? We’ll integrate code LLaMA into Perplexity, all in service of providing you with the best answers to your most technical questions!
0:12

Code LLaMA is now on Perplexity’s LLaMa Chat! Try asking it to write a function for you, or explain a code snippet: 🔗 This is the fastest way to try AI at Meta’s latest code-specialized LLM. With our model deployment expertise, we are able to provide you with this model less than 24 hours of it’s release. What’s next? We’ll integrate code LLaMA into Perplexity, all in service of providing you with the best answers to your most technical questions!

Perplexity

274,587 views • 3 years ago

There is a volcano erupting in Hawaii, but because of the government shutdown the agency tasked with tracking it hasn’t been able to share the news on social media. 🌋 Fortunately no one is in danger. ⚠️
0:13

There is a volcano erupting in Hawaii, but because of the government shutdown the agency tasked with tracking it hasn’t been able to share the news on social media. 🌋 Fortunately no one is in danger. ⚠️

DramaAlert

160,643 views • 11 months ago

Whatever you’re building, a game, an animation, or a world of your own, 404 helps you turn ideas into usable 3D assets. Build beautiful organic worlds using 404-generated assets. Every asset in this reel was generated with 404. All you bring is the prompt. Try it out at:
0:15

Whatever you’re building, a game, an animation, or a world of your own, 404 helps you turn ideas into usable 3D assets. Build beautiful organic worlds using 404-generated assets. Every asset in this reel was generated with 404. All you bring is the prompt. Try it out at:

404

15,327 views • 1 month ago