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。

Announcing FPV-SLAM, a state-of-the-art SLAM system for egocentric data. FPV-SLAM ranks 1st in the monocular-inertial track of LaMAria, ETH Zurich's city-scale benchmark.

Abhishek Anand

6,380 subscribers

13,647 次观看 • 12 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

30 条评论

Abhishek Anand 的头像

Abhishek Anand12 天前

Read the full blog here:

Satpal Singh Rathore 的头像

Satpal Singh Rathore12 天前

@fpv_labs Leading across all categories of short, med, long, low light, moving platform etc.

Gunjan Paul 的头像

Gunjan Paul12 天前

@pratyush21225 @fpv_labs Congrats!!! No. 1 in LaMAria is crazy 🚀🚀

Abhishek Anand 的头像

Abhishek Anand12 天前

@pratyush21225 @fpv_labs Thank you sir!

Asad Memon 的头像

Asad Memon11 天前

@fpv_labs Congrats on the lead! Looking forward to reading about it (and improving Mighty Camera) 😬

Annu Shekhawat 的头像

Annu Shekhawat12 天前

@fpv_labs Congratulations to the whole team! 🌟🔥

Abhishek Anand 的头像

Abhishek Anand12 天前

@fpv_labs Thanks - we've come a long way since we published our first blog on the challenges with egocentric SLAM

Shubhanshu Khatana 的头像

Shubhanshu Khatana12 天前

@fpv_labs LFG!!!🚀🚀🚀🚀

Abhishek Anand 的头像

Abhishek Anand12 天前

@fpv_labs LFG!

ekaksh janweja 的头像

ekaksh janweja12 天前

@fpv_labs congratulations 🔥

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs Thanks buddy!

Sourik 的头像

Sourik12 天前

@fpv_labs Congratulations!

Abhishek Anand 的头像

Abhishek Anand12 天前

@fpv_labs Thank you sir!

yogi 的头像

yogi12 天前

@fpv_labs congrats on the launch! lfgg

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs ♥️

mansin 的头像

mansin12 天前

@fpv_labs Lfg!

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs LFG - we need to catchup soon!

Pratyush Patnaik 的头像

Pratyush Patnaik12 天前

@fpv_labs lfg!! 🚀

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs LFGG!

Esteban Damian Fuhrmann 的头像

Esteban Damian Fuhrmann11 天前

@fpv_labs Really impressive work! 🔥 Have you tested FPV-SLAM on long egocentric videos with lots of motion blur? I’d love to try it with our camera + IMU data!

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs sure can you pls email me at [email protected] - will connect with team who can help!

Anindyadeep 的头像

Anindyadeep12 天前

@fpv_labs Lfgg congrats on the launch

Abhishek Anand 的头像

Abhishek Anand12 天前

@fpv_labs Thanks buddy!

F-G Fernandez 的头像

F-G Fernandez12 天前

@fpv_labs Congrats, mono+IMU on head-worn footage is the hard track: fast head turns, motion blur, people moving all around you. Is the ranked run online, or offline with full-sequence optimization? Both are useful, but they suit different products.

Srijan Jain 的头像

Srijan Jain12 天前

@fpv_labs Congratulations!

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs Thanks sir!

vrushtee 的头像

vrushtee10 天前

@fpv_labs congratss!!

Abhishek Anand 的头像

Abhishek Anand10 天前

@fpv_labs Thank youu!

Mahmood Ul Hassan, PhD 的头像

Mahmood Ul Hassan, PhD10 天前

@fpv_labs City-scale monocular-inertial SLAM is a serious challenge! Impressive benchmark result for FPV-SLAM. Robust localization and loop closure in complex environments are key to scalable autonomous navigation.

Bruno Santos🇵🇹 的头像

Bruno Santos🇵🇹10 天前

@fpv_labs Good job !

相关视频

Today, we are announcing GroundedSLAM. LaMAria is the most difficult egocentric SLAM benchmark to date. We are the first to close the gap on Aria’s SLAM and eclipse all other methods by a wide margin. SLAM is the foundation of all 3D estimation. We think this is a big unlock for better robotics 👇🧵

Today, we are announcing GroundedSLAM. LaMAria is the most difficult egocentric SLAM benchmark to date. We are the first to close the gap on Aria’s SLAM and eclipse all other methods by a wide margin. SLAM is the foundation of all 3D estimation. We think this is a big unlock for better robotics 👇🧵

Grounded Superintelligence

101,416 次观看 • 1 个月前

Today we're announcing microSLAM, a monocular SLAM system built by our computer vision team in Zürich with ETH Zurich's Computer Vision and Geometry Lab. It ranks first on the LaMaria benchmark for monocular SLAM, and holds up against systems that carry many more cameras and IMUs. Ours runs on a single RGB stream. Robot learning is bottlenecked on data, and the largest untapped source is people going about their work. A camera on someone's head for an afternoon in a plant is a record of how that plant actually operates, including all the parts nobody writes down. That data is only useful if you can recover the geometry, and geometry from one moving camera in a world that will not sit still is the hard version of the problem. Solving it monocular is what makes the data cheap. Rigs do not scale to every worker. Glasses do. microSLAM turns that footage into interactive 3D environments where behavioral foundation models can be trained. One person walks a factory floor. A robot learns to navigate it.

Today we're announcing microSLAM, a monocular SLAM system built by our computer vision team in Zürich with ETH Zurich's Computer Vision and Geometry Lab. It ranks first on the LaMaria benchmark for monocular SLAM, and holds up against systems that carry many more cameras and IMUs. Ours runs on a single RGB stream. Robot learning is bottlenecked on data, and the largest untapped source is people going about their work. A camera on someone's head for an afternoon in a plant is a record of how that plant actually operates, including all the parts nobody writes down. That data is only useful if you can recover the geometry, and geometry from one moving camera in a world that will not sit still is the hard version of the problem. Solving it monocular is what makes the data cheap. Rigs do not scale to every worker. Glasses do. microSLAM turns that footage into interactive 3D environments where behavioral foundation models can be trained. One person walks a factory floor. A robot learns to navigate it.

Gianni Hodel

76,090 次观看 • 2 个月前

All ready for the first Slam of the year! Rekortan 🤝 Grand Slam Track

All ready for the first Slam of the year! Rekortan 🤝 Grand Slam Track

Grand Slam Track

34,561 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.

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

Michael Johnson unveils new Track League to be called Grand Slam Track! - $12.6 million prize money in total - $100k for the winner of each slam - 4 Annual slams - Each slam is 3 days long - 96 athletes - No field events - No pacemakers

Michael Johnson unveils new Track League to be called Grand Slam Track! - $12.6 million prize money in total - $100k for the winner of each slam - 4 Annual slams - Each slam is 3 days long - 96 athletes - No field events - No pacemakers

Track & Field Gazette

172,346 次观看 • 2 年前

We are publishing our first deep dive on what we believe is one of the most challenging layers in egocentric data - SLAM and VIO in the context of long-horizon state tracking. We break down how SLAM and VIO fail in egocentric settings - visual features vanish at close range, depth sensors saturate, fast head motion blurs frames, and these failures don't always occur in isolation. They hit at the exact same moment, leading to compounding errors and making the downstream data unusable. We believe the foundation for high-quality egocentric data demands sub-centimeter precision over long episodes ranging from a few minutes to up to an hour.

We are publishing our first deep dive on what we believe is one of the most challenging layers in egocentric data - SLAM and VIO in the context of long-horizon state tracking. We break down how SLAM and VIO fail in egocentric settings - visual features vanish at close range, depth sensors saturate, fast head motion blurs frames, and these failures don't always occur in isolation. They hit at the exact same moment, leading to compounding errors and making the downstream data unusable. We believe the foundation for high-quality egocentric data demands sub-centimeter precision over long episodes ranging from a few minutes to up to an hour.

FPV Labs

42,211 次观看 • 7 个月前

GIANNIS SLAM. GIANNIS SLAM. GIANNIS SLAM. THREE DUNKS IN A ROW FOR THE GREEK FREAK 💥💥💥

GIANNIS SLAM. GIANNIS SLAM. GIANNIS SLAM. THREE DUNKS IN A ROW FOR THE GREEK FREAK 💥💥💥

NBA

331,501 次观看 • 10 个月前

🏆 NSAA TOURNAMENT UPDATE 🚨🚨DAKOTA STATE 2-OUT GRAND SLAM FOR THE LEAD Trojans 3B Dawson Portner delivers a go-ahead grand slam in the top of the 9th to extend the season for #5 DSU in a 5-4 victory over #8 Valley City State City State! Sam Tyrpa 4.2 IP 0 ER 7 K in relief DSU will take on the loser of #3 Dickinson State/ #7 Mayville State tomorrow 🎥: North Star #NAIABall

🏆 NSAA TOURNAMENT UPDATE 🚨🚨DAKOTA STATE 2-OUT GRAND SLAM FOR THE LEAD Trojans 3B Dawson Portner delivers a go-ahead grand slam in the top of the 9th to extend the season for #5 DSU in a 5-4 victory over #8 Valley City State City State! Sam Tyrpa 4.2 IP 0 ER 7 K in relief DSU will take on the loser of #3 Dickinson State/ #7 Mayville State tomorrow 🎥: North Star #NAIABall

NAIA Ball

17,569 次观看 • 1 年前

The threat of the slam can be as strong as the slam itself

The threat of the slam can be as strong as the slam itself

BM | Ticklememour @ Sandia Showdown 2026!!!!!

72,930 次观看 • 1 年前

Tom Hamilton’s Call of Carlos Santana’s GRAND SLAM in the GRAND SLAM GIVEAWAY INNING #GuardsBall

Tom Hamilton’s Call of Carlos Santana’s GRAND SLAM in the GRAND SLAM GIVEAWAY INNING #GuardsBall

k

61,192 次观看 • 1 年前

Malcolm Moore GRAND SLAM 💥 The 2024 1st rounder is hitting .360 with a 1.027 OPS for Hub City Spartanburgers in May

Malcolm Moore GRAND SLAM 💥 The 2024 1st rounder is hitting .360 with a 1.027 OPS for Hub City Spartanburgers in May

Rangers Player Development

35,573 次观看 • 6 个月前

𝑪𝑨𝑳𝑳𝑰𝑵𝑮 𝑨𝑳𝑳 𝑹𝑰𝑽𝑨𝑳𝑺. This is the future of track. This is Grand Slam Track. Get in the Race: #GrandSlamTrack #OnlyTheFastest

𝑪𝑨𝑳𝑳𝑰𝑵𝑮 𝑨𝑳𝑳 𝑹𝑰𝑽𝑨𝑳𝑺. This is the future of track. This is Grand Slam Track. Get in the Race: #GrandSlamTrack #OnlyTheFastest

Grand Slam Track

154,862 次观看 • 2 年前

Monocular visual slam with loop closures on a $10 chip. cc Akshat

Monocular visual slam with loop closures on a $10 chip. cc Akshat

mansin

17,622 次观看 • 1 个月前

SEE YOU SOON, PHILLY 👊 Grand Slam Track heads to the City of Brotherly Love from May 30-June 1, 2025.

SEE YOU SOON, PHILLY 👊 Grand Slam Track heads to the City of Brotherly Love from May 30-June 1, 2025.

Grand Slam Track

16,642 次观看 • 2 年前