正在加载视频...

视频加载失败

返回首页

Camera frame registration with little or no overlap? Check out Can Gümeli's ObjectMatch: We leverage object semantics to find indirect correspondences between frames. Great for any SLAM and SfM pose optimization! Vid: #CVPR2023 Angela Dai

MattNiessner's profile picture

Matthias Niessner

48,723 subscribers

17,749 次观看 • 3 年前 •via X (Twitter)

#CVPR2023
LIVE

Anya Rossi• Live Now

Private livecam show

0 条评论

暂无评论

原始帖子的评论将显示在这里

相关视频

📢Face Anything: 4D Face Reconstruction from Any Image Sequence Transformer model for 4D face reconstruction and dense tracking: - predict canonical facial coordinates per pixel - tracking as reconstruction in canonical space - geometry + correspondences in one forward pass Key idea: a shared canonical space across frames - correspondences as nearest neighbors - no motion or deformation estimation Stable geometry and tracking, even under large expressions and viewpoint changes - check out our results! 🌐 ▶️ Great work by Umut Kocasarı, Simon Giebenhain, Richard Shaw
0:42

📢Face Anything: 4D Face Reconstruction from Any Image Sequence Transformer model for 4D face reconstruction and dense tracking: - predict canonical facial coordinates per pixel - tracking as reconstruction in canonical space - geometry + correspondences in one forward pass Key idea: a shared canonical space across frames - correspondences as nearest neighbors - no motion or deformation estimation Stable geometry and tracking, even under large expressions and viewpoint changes - check out our results! 🌐 ▶️ Great work by Umut Kocasarı, Simon Giebenhain, Richard Shaw

Matthias Niessner

61,705 次观看 • 3 个月前

Drop the camera, and bang -- the robots do the job without camera calibration or additional data collection. We want robots to manipulate anywhere! Our new work, **Maniwhere**, enables robots to manipulate objects from any camera view with any background. Isn't this the generalization every roboticist is looking for? How do we achieve this with sim2real? You can check this thread from Zhecheng Yuan ! And the website is here:
0:29

Drop the camera, and bang -- the robots do the job without camera calibration or additional data collection. We want robots to manipulate anywhere! Our new work, **Maniwhere**, enables robots to manipulate objects from any camera view with any background. Isn't this the generalization every roboticist is looking for? How do we achieve this with sim2real? You can check this thread from Zhecheng Yuan ! And the website is here:

Huazhe Harry Xu

27,564 次观看 • 2 年前

Can we scale 4D pretraining to learn general space-time representations that reconstruct an object from a few views at any time to any view at any other time? Introducing 4D-LRM: a Large Space-Time Reconstruction Model that ... 🔹 Predicts 4D Gaussian primitives directly from multi-view tokens (no motion vectors, no HexPlane); 🔹 Uses a clean, minimal Transformer backbone; 🔹 Generalizes with fast, high-quality feedforward rendering at any view and infinite frame rate. Check out more interactive demos and scaling behaviors on our homepage/paper. 👉Website: 👉Paper:
2:30

Can we scale 4D pretraining to learn general space-time representations that reconstruct an object from a few views at any time to any view at any other time? Introducing 4D-LRM: a Large Space-Time Reconstruction Model that ... 🔹 Predicts 4D Gaussian primitives directly from multi-view tokens (no motion vectors, no HexPlane); 🔹 Uses a clean, minimal Transformer backbone; 🔹 Generalizes with fast, high-quality feedforward rendering at any view and infinite frame rate. Check out more interactive demos and scaling behaviors on our homepage/paper. 👉Website: 👉Paper:

Martin Ziqiao Ma

21,811 次观看 • 1 年前

You can now use SUKU to pay for gas in any DApp⛽ Great news! Migration to Account Abstraction is complete and the SukuWallet’s biggest update yet is here! We are super thrilled to share it with you all. Check it out 🧵👇
0:22

You can now use SUKU to pay for gas in any DApp⛽ Great news! Migration to Account Abstraction is complete and the SukuWallet’s biggest update yet is here! We are super thrilled to share it with you all. Check it out 🧵👇

Suku

31,975 次观看 • 1 年前

This seemingly obvious prediction didn't take long to become reality. MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors Contributions: • The first real-time SLAM system using the two-view 3D reconstruction prior MASt3R [20] as a foundation. • Efficient techniques for pointmap matching, tracking and local fusion, graph construction and loop closure, and second-order global optimization. • A state-of-the-art dense SLAM system capable of handling generic, time-varying camera models. Abstract: We present a real-time monocular dense SLAM system, designed from the ground up using MASt3R, a two-view 3D reconstruction and matching prior. Equipped with this strong prior, our system remains robust on in-the-wild video sequences, making no assumptions on a fixed or parametric camera model beyond a unique camera center. Key features include: - Efficient methods for pointmap matching, camera tracking, and local fusion - Graph construction and loop closure - Second-order global optimization With known calibration, a simple modification achieves state-of-the-art performance across various benchmarks. Altogether, we propose a plug-and-play monocular SLAM system capable of producing globally-consistent poses and dense geometry while operating at 15 FPS.
0:23

This seemingly obvious prediction didn't take long to become reality. MASt3R-SLAM: Real-Time Dense SLAM with 3D Reconstruction Priors Contributions: • The first real-time SLAM system using the two-view 3D reconstruction prior MASt3R [20] as a foundation. • Efficient techniques for pointmap matching, tracking and local fusion, graph construction and loop closure, and second-order global optimization. • A state-of-the-art dense SLAM system capable of handling generic, time-varying camera models. Abstract: We present a real-time monocular dense SLAM system, designed from the ground up using MASt3R, a two-view 3D reconstruction and matching prior. Equipped with this strong prior, our system remains robust on in-the-wild video sequences, making no assumptions on a fixed or parametric camera model beyond a unique camera center. Key features include: - Efficient methods for pointmap matching, camera tracking, and local fusion - Graph construction and loop closure - Second-order global optimization With known calibration, a simple modification achieves state-of-the-art performance across various benchmarks. Altogether, we propose a plug-and-play monocular SLAM system capable of producing globally-consistent poses and dense geometry while operating at 15 FPS.

MrNeRF

29,974 次观看 • 1 年前

📢WorldAgents: 3D worlds only from 2D image models - without any training! We propose an agentic approach with a Director (VLM) to plan the scene, a Generator (Flux or NanoBanana) for new views, and a Verifier (VLM) for selection / 3D consistency. -> High-fidelity 3D worlds from a single text prompt. What's remarkable: our agents find consistent views from 2D image models to obtain 3D-consistent worlds; this shows that image models contain world priors - agents just need to find them! Great work by Ziya Erkoç Angela Dai
0:25

📢WorldAgents: 3D worlds only from 2D image models - without any training! We propose an agentic approach with a Director (VLM) to plan the scene, a Generator (Flux or NanoBanana) for new views, and a Verifier (VLM) for selection / 3D consistency. -> High-fidelity 3D worlds from a single text prompt. What's remarkable: our agents find consistent views from 2D image models to obtain 3D-consistent worlds; this shows that image models contain world priors - agents just need to find them! Great work by Ziya Erkoç Angela Dai

Matthias Niessner

19,005 次观看 • 4 个月前

We made an interactive client-server viewer for LagerNVS with Jonathon Luiten! You can now interactively explore scenes from just a photo capture - no optimization, no 3D Gaussians, just load your images, run the model on a cloud GPU and stream the renders to your local browser. Check out the video below for some spaces I recently captured in Oxford, London and beyond!
1:38

We made an interactive client-server viewer for LagerNVS with Jonathon Luiten! You can now interactively explore scenes from just a photo capture - no optimization, no 3D Gaussians, just load your images, run the model on a cloud GPU and stream the renders to your local browser. Check out the video below for some spaces I recently captured in Oxford, London and beyond!

Stan Szymanowicz

16,850 次观看 • 3 个月前

What a spot for a trail camera! We put this camera out on a small footpath this winter for the first time just out of curiosity. We do a lot of trial and error to find good spots that wolves like to use often. The little trail is pretty close to the intersection of 3 different pack territories and we figured it might be a good place…and it was. We captured more footage of wolves and other animals on this camera that are not in this video. This video is just the highlights. Most of the other videos are at night. Notably, for every camera like this that gets great footage, we probably have 4 or 5 that are duds…and since they are duds, we don’t share that footage online!
1:52

What a spot for a trail camera! We put this camera out on a small footpath this winter for the first time just out of curiosity. We do a lot of trial and error to find good spots that wolves like to use often. The little trail is pretty close to the intersection of 3 different pack territories and we figured it might be a good place…and it was. We captured more footage of wolves and other animals on this camera that are not in this video. This video is just the highlights. Most of the other videos are at night. Notably, for every camera like this that gets great footage, we probably have 4 or 5 that are duds…and since they are duds, we don’t share that footage online!

Voyageurs Wolf Project

21,477 次观看 • 3 个月前

Yesterday: had juice. Felt great. Today: minimal juice. No caffeine. No training partner To expect that you’ll be “on” every day is either a great way to utilize delusion to create energy, or set yourself up for disappointing performance. Tough days training or in game are a signal. That’s it. You can ruminate on them with no changed course of action OR You can find out WHY you’re not up to standard. In this case, it’s sleep, carb intake, and managing stressors (for me personally) Be intuitive. The strap or the ring can only tell you so much.
0:30

Yesterday: had juice. Felt great. Today: minimal juice. No caffeine. No training partner To expect that you’ll be “on” every day is either a great way to utilize delusion to create energy, or set yourself up for disappointing performance. Tough days training or in game are a signal. That’s it. You can ruminate on them with no changed course of action OR You can find out WHY you’re not up to standard. In this case, it’s sleep, carb intake, and managing stressors (for me personally) Be intuitive. The strap or the ring can only tell you so much.

Jack Barry

11,012 次观看 • 2 年前

Whilst we figure out a solution for temporal filtering on 3D Gaussian Splatting input frames. We can use a temporal filtering plugin on the screen captures or rendered images to act as a visualization tool to check what the future results might look like. "Processing and rendering a digital human with 3D Gaussian Splatting with a post process to simulate temporal filtering." Left video no temporal filtering. Right video with temporal filtering. R&D for archviz projects. #GaussianSplatting #ir #inria #sibr #aftereffects #temporalfiltering
0:42

Whilst we figure out a solution for temporal filtering on 3D Gaussian Splatting input frames. We can use a temporal filtering plugin on the screen captures or rendered images to act as a visualization tool to check what the future results might look like. "Processing and rendering a digital human with 3D Gaussian Splatting with a post process to simulate temporal filtering." Left video no temporal filtering. Right video with temporal filtering. R&D for archviz projects. #GaussianSplatting #ir #inria #sibr #aftereffects #temporalfiltering

Infinite-Realities

58,383 次观看 • 2 年前

Android 17 is rolling out to Pixel devices today! 🚀 Huge congrats to the Android team! Here are some of my favorite highlights: 🫧 Bubbles: Long-press any app to turn it into a floating window or dock it in a bottom bubble bar on large screens to seamlessly multitask. 🤳 Screen Reactions: Simultaneously record your screen and your selfie camera to easily share video reactions over apps or sites. 🎮 Foldable Gaming: Level up with an optimized 50/50 split layout, native controller remapping, and memory optimization to stop frame drops. 🔒 Security Safeguards: Take control with temporary location access, selective contact sharing, and biometric lockdown in Find Hub. Look out for the update on your Pixel device, and other Android devices throughout 2026. 👇 Here's Dieter Bohn explaining Bubbles!
0:46

Android 17 is rolling out to Pixel devices today! 🚀 Huge congrats to the Android team! Here are some of my favorite highlights: 🫧 Bubbles: Long-press any app to turn it into a floating window or dock it in a bottom bubble bar on large screens to seamlessly multitask. 🤳 Screen Reactions: Simultaneously record your screen and your selfie camera to easily share video reactions over apps or sites. 🎮 Foldable Gaming: Level up with an optimized 50/50 split layout, native controller remapping, and memory optimization to stop frame drops. 🔒 Security Safeguards: Take control with temporary location access, selective contact sharing, and biometric lockdown in Find Hub. Look out for the update on your Pixel device, and other Android devices throughout 2026. 👇 Here's Dieter Bohn explaining Bubbles!

Sameer Samat

25,641 次观看 • 2 个月前

Do you think you and your friend can take on any challenge? We'd like to see you try with our new bite-sized series! Play We Were Here Expeditions: The FriendShip and find out if you will sink or swim together. 💥 Claim the game for FREE until Oct 13! 💥
2:08

Do you think you and your friend can take on any challenge? We'd like to see you try with our new bite-sized series! Play We Were Here Expeditions: The FriendShip and find out if you will sink or swim together. 💥 Claim the game for FREE until Oct 13! 💥

We Were Here Series

91,469 次观看 • 2 年前

We are completely exhausted… we can no longer bear more. Everything feels heavier with each passing day. We try to hold on, we try to be patient and stay strong, but what we live through here in Gaza every single day is beyond what any human being can endure. Seeing these streets, this harsh reality, and the difficult life we are forced to live day after day is breaking us little by little. Please, just support us. Sometimes, a single word, a single share, or a single prayer can change everything… For Rafif… for my family.
1:04

We are completely exhausted… we can no longer bear more. Everything feels heavier with each passing day. We try to hold on, we try to be patient and stay strong, but what we live through here in Gaza every single day is beyond what any human being can endure. Seeing these streets, this harsh reality, and the difficult life we are forced to live day after day is breaking us little by little. Please, just support us. Sometimes, a single word, a single share, or a single prayer can change everything… For Rafif… for my family.

Mohammed jawad 🇵🇸

19,515 次观看 • 1 个月前

This Google Veo 3 "frames to video" feature is pure genius 💡 If you're creating video ads, this Veo 3 technique gives you surgical control over every scene element. This frames-to-video feature will take any static product shot... And turn it into cinematic reels with precise camera movements & timing. Perfect for creative strategists, agencies, and DTC operators running video campaigns. But instead of hoping AI generates what you hope for... Just upload your reference image and get exact scene control in minutes. The game-changer? Your static assets become dynamic video gold → organized by camera moves, action sequences, & precise timing control. Here's what frames-to-video gives you: → upload any product reference image → add numbered annotations for camera movements → structure scene directions (I use JSON for precision) → set duration, fps, lighting specs → get frame-perfect video execution → no video shooting or expensive production needed Now you've got director-level control from single images. These HexClad reels were generated from two static product shots + a JSON prompt for each. When your campaign needs cinematic quality but you only have product photos? This frames-to-video workflow changes everything. And it's only getting better Want my full annotation setup guide, including the JSON prompts? Comment "FRAMES" + follow and I'll send it over (must be following/connected so I can DM)
0:32

This Google Veo 3 "frames to video" feature is pure genius 💡 If you're creating video ads, this Veo 3 technique gives you surgical control over every scene element. This frames-to-video feature will take any static product shot... And turn it into cinematic reels with precise camera movements & timing. Perfect for creative strategists, agencies, and DTC operators running video campaigns. But instead of hoping AI generates what you hope for... Just upload your reference image and get exact scene control in minutes. The game-changer? Your static assets become dynamic video gold → organized by camera moves, action sequences, & precise timing control. Here's what frames-to-video gives you: → upload any product reference image → add numbered annotations for camera movements → structure scene directions (I use JSON for precision) → set duration, fps, lighting specs → get frame-perfect video execution → no video shooting or expensive production needed Now you've got director-level control from single images. These HexClad reels were generated from two static product shots + a JSON prompt for each. When your campaign needs cinematic quality but you only have product photos? This frames-to-video workflow changes everything. And it's only getting better Want my full annotation setup guide, including the JSON prompts? Comment "FRAMES" + follow and I'll send it over (must be following/connected so I can DM)

Mike Futia

39,838 次观看 • 1 年前

Fun fact: It would take the average PC roughly 138,069,000 hours or about 15,761 years to render Avatar: The Way of Water. Here’s how I worked it out: The average PC in 2024 has 6 cores. The average blade on WetaFXs renderfarm has between 16-64 cores. There is roughly 414,000 frames in the finished film, accounting for half of the film being 48fps and half being 24fps at the final runtime of 192 minutes. Each frame took an average of 20-80 hours to render on WetaFXs blades depending on shot complexity and accounting for the 4K delivery resolution. With that info, we can work out the average: We say that the average Weta blade has 40 cores (average between the 16-64 count). That gives a Weta to PC core ratio of 6.67 We also say that the average render time per frame is 50 hours (average of 20-80) Total render time for Weta: 50 hours x 414,000 frames = 20,700,000 hours To work out how long this would be on the average home PC we must multiply by the core ratio from earlier: 50 hours x 6.67 = 333.5 hours per frame Total render time on a home PC: 333.5 hours x 414,000 frames = 138,069,000 hours, which converted to years is 15,761 DISCLAIMER: This is all an educated guess done for fun based on publicly available technical information about the film and Wetas renderfarm. It does not account for any production optimizations, storage limits etc. The real number could be drastically higher. CREDITS: Video is from WetaFXs YouTube channel:
1:49

Fun fact: It would take the average PC roughly 138,069,000 hours or about 15,761 years to render Avatar: The Way of Water. Here’s how I worked it out: The average PC in 2024 has 6 cores. The average blade on WetaFXs renderfarm has between 16-64 cores. There is roughly 414,000 frames in the finished film, accounting for half of the film being 48fps and half being 24fps at the final runtime of 192 minutes. Each frame took an average of 20-80 hours to render on WetaFXs blades depending on shot complexity and accounting for the 4K delivery resolution. With that info, we can work out the average: We say that the average Weta blade has 40 cores (average between the 16-64 count). That gives a Weta to PC core ratio of 6.67 We also say that the average render time per frame is 50 hours (average of 20-80) Total render time for Weta: 50 hours x 414,000 frames = 20,700,000 hours To work out how long this would be on the average home PC we must multiply by the core ratio from earlier: 50 hours x 6.67 = 333.5 hours per frame Total render time on a home PC: 333.5 hours x 414,000 frames = 138,069,000 hours, which converted to years is 15,761 DISCLAIMER: This is all an educated guess done for fun based on publicly available technical information about the film and Wetas renderfarm. It does not account for any production optimizations, storage limits etc. The real number could be drastically higher. CREDITS: Video is from WetaFXs YouTube channel:

Rassoul Edji

37,819 次观看 • 1 年前

📢 OneCanvas: 3D Scene Understanding via Panoramic Reprojection We extract features from video frames and reproject them into one occlusion-free view of the whole scene that a 2D VLM reads just like a normal image. We can center this view on any viewpoint, including an agent's own pose for situated reasoning. The same projection lets us create spatial training tasks with no human annotation, solvable only by reasoning over the 3D positions of real object features placed on an otherwise empty canvas. The result is a stock 2D VLM that reasons in 3D, setting a new state of the art across spatial benchmarks at far less compute. 🌐 ▶️ Great work by Bartłomiej Baranowski & Dave Zhenyu Chen
0:44

📢 OneCanvas: 3D Scene Understanding via Panoramic Reprojection We extract features from video frames and reproject them into one occlusion-free view of the whole scene that a 2D VLM reads just like a normal image. We can center this view on any viewpoint, including an agent's own pose for situated reasoning. The same projection lets us create spatial training tasks with no human annotation, solvable only by reasoning over the 3D positions of real object features placed on an otherwise empty canvas. The result is a stock 2D VLM that reasons in 3D, setting a new state of the art across spatial benchmarks at far less compute. 🌐 ▶️ Great work by Bartłomiej Baranowski & Dave Zhenyu Chen

Matthias Niessner

25,198 次观看 • 1 个月前

Shortly before midnight, we had multiple reports sent in of an extremely bright green flash in the sky to the north. We went back in our database of Aurora cameras to find the culprit… and here it is!! The Manistee camera had the best angle as the object streaked across the northern sky. You can even pick up on a faint auroral glow at the same time! Not only that, but you can see the streak left behind from the flash after it occurred. Time was 11:13 PM EDT and approximately 35 seconds, October 29th, 2025, for those who’d also like to check any home security cameras.
0:21

Shortly before midnight, we had multiple reports sent in of an extremely bright green flash in the sky to the north. We went back in our database of Aurora cameras to find the culprit… and here it is!! The Manistee camera had the best angle as the object streaked across the northern sky. You can even pick up on a faint auroral glow at the same time! Not only that, but you can see the streak left behind from the flash after it occurred. Time was 11:13 PM EDT and approximately 35 seconds, October 29th, 2025, for those who’d also like to check any home security cameras.

Michigan Storm Chasers

19,421 次观看 • 9 个月前

Kevin O'Leary explains why he avoids hiring workaholics. "I don't hire workaholics. If you look at great managers, great investors, great entrepreneurs, you're going to find a common trait, what they do with their spare time. And you're going to find out it's something in the arts or it's something in sports, something that has nothing to do with business. Because you need the balance in your brain between the chaos of the arts and the discipline of business." PS. We post daily content strictly for dedicated entrepreneurs, so if you’re one of them, make sure to follow us Entrepreneurs on X for more. Like and repost if you found value in this post:
0:22

Kevin O'Leary explains why he avoids hiring workaholics. "I don't hire workaholics. If you look at great managers, great investors, great entrepreneurs, you're going to find a common trait, what they do with their spare time. And you're going to find out it's something in the arts or it's something in sports, something that has nothing to do with business. Because you need the balance in your brain between the chaos of the arts and the discipline of business." PS. We post daily content strictly for dedicated entrepreneurs, so if you’re one of them, make sure to follow us Entrepreneurs on X for more. Like and repost if you found value in this post:

Entrepreneurs on X

33,027 次观看 • 1 个月前