Video wird geladen...

Video konnte nicht geladen werden

Beim Laden dieses Videos ist ein Problem aufgetreten. Dies könnte an einem vorübergehenden Netzwerkproblem liegen oder das Video ist möglicherweise nicht verfügbar.

open_semantic_slam ICRA2025: OpenGS-SLAM: Open-Set Dense Semantic SLAM with 3D Gaussian Splatting for Object-Level Scene Understanding

rsasaki0109

6,931 subscribers

27,137 Aufrufe • vor 1 Jahr •via X (Twitter)

Bildung Wissenschaft & Technologie

Anya Rossi• Live Now

Private livecam show

3 Kommentare

Profilbild von Anto Patrex

Anto Patrexvor 1 Jahr

Is this realtime?

Profilbild von Rainmaker

Rainmakervor 1 Jahr

Can reinforcement learning handle stock market swings? In my latest free Substack, find out how SARSA reinforcement learning algorithm can help create adaptive strategies and improve performance.

Profilbild von Albert

Albertvor 1 Jahr

how about in dynamic environments and under heavy occlusions?"

Ähnliche Videos

LEGO-SLAM: Language-Embedded Gaussian Optimization SLAM LEGO-SLAM running at 15 FPS on a ScanNet scene with language-based loop closing for drift correction. LEGO-SLAM is a 3DGS-based SLAM framework that supports open-vocabulary semantic querying and rendering. It tracks via G-ICP and efficiently builds a map by embedding Gaussians with scene-adaptive 16D language features. Map management is achieved through Language Pruning and Language-Based Loop Detection. The generated map enables open-vocabulary 3D Object Localization.

LEGO-SLAM: Language-Embedded Gaussian Optimization SLAM LEGO-SLAM running at 15 FPS on a ScanNet scene with language-based loop closing for drift correction. LEGO-SLAM is a 3DGS-based SLAM framework that supports open-vocabulary semantic querying and rendering. It tracks via G-ICP and efficiently builds a map by embedding Gaussians with scene-adaptive 16D language features. Map management is achieved through Language Pruning and Language-Based Loop Detection. The generated map enables open-vocabulary 3D Object Localization.

Ryohei Sasaki@engineer

14,935 Aufrufe • vor 3 Monaten

"Triangle Splatting SLAM" TL;DR: the first dense RGB-D SLAM system built on differentiable triangle splatting, enabling real-time tracking, mesh reconstruction, collision checking, and scene editing from a unified representation.

"Triangle Splatting SLAM" TL;DR: the first dense RGB-D SLAM system built on differentiable triangle splatting, enabling real-time tracking, mesh reconstruction, collision checking, and scene editing from a unified representation.

Alexandre Morgand

11,804 Aufrufe • vor 21 Tagen

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

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

MrNeRF

35,669 Aufrufe • vor 2 Jahren

Code released: "GaussianObject: Just Taking Four Images to Get A High-Quality 3D Object with Gaussian Splatting"

Code released: "GaussianObject: Just Taking Four Images to Get A High-Quality 3D Object with Gaussian Splatting"

MrNeRF

22,274 Aufrufe • vor 2 Jahren

SuperGSeg: Open-Vocabulary 3D Segmentation with Structured Super-Gaussians Contributions: • We propose SuperGSeg: a 3D segmentation method with neural Gaussians, designed to learn hierarchical instance segmentation features from 2D foundation models. • We introduce the concept of Super-Gaussian, a novel representation that integrates hierarchical instance segmentation features, enabling the embedding of high-dimensional language features. This approach addresses previously unfeasible challenges in representing complex scenes with rich semantic details. • Extensive experiments on the LERF-OVS and ScanNet datasets demonstrate the effectiveness of the proposed method, achieving significant improvements in open-vocabulary 3D object-level and scene-level semantic segmentation. It shows particular strength in capturing fine-grained scene details and dense pixel semantic segmentation tasks for the first time.

SuperGSeg: Open-Vocabulary 3D Segmentation with Structured Super-Gaussians Contributions: • We propose SuperGSeg: a 3D segmentation method with neural Gaussians, designed to learn hierarchical instance segmentation features from 2D foundation models. • We introduce the concept of Super-Gaussian, a novel representation that integrates hierarchical instance segmentation features, enabling the embedding of high-dimensional language features. This approach addresses previously unfeasible challenges in representing complex scenes with rich semantic details. • Extensive experiments on the LERF-OVS and ScanNet datasets demonstrate the effectiveness of the proposed method, achieving significant improvements in open-vocabulary 3D object-level and scene-level semantic segmentation. It shows particular strength in capturing fine-grained scene details and dense pixel semantic segmentation tasks for the first time.

MrNeRF

13,594 Aufrufe • vor 1 Jahr

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

"YoNoSplat: You Only Need One Model for Feedforward 3D Gaussian Splatting" TL;DR: a unified 3D Gaussian splatting model that reconstructs high-quality scene geometry and camera poses from unposed/uncalibrated images in a single forward pass.

Alexandre Morgand

14,839 Aufrufe • vor 4 Monaten

LangSplatV2 High-dimensional 3D Language Gaussian Splatting with 450+ FPS

LangSplatV2 High-dimensional 3D Language Gaussian Splatting with 450+ FPS

AK

46,826 Aufrufe • vor 11 Monaten

WaterSplatting: Fast Underwater 3D Scene Reconstruction Using Gaussian Splatting Project: Method ⬇️ 1 I 2

WaterSplatting: Fast Underwater 3D Scene Reconstruction Using Gaussian Splatting Project: Method ⬇️ 1 I 2

MrNeRF

12,175 Aufrufe • vor 1 Jahr

VolSplat Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction

VolSplat Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction

AK

17,660 Aufrufe • vor 9 Monaten

Goodbye SOGS, hello SOG! 👋 PlayCanvas open sources Spatially Ordered Gaussians - a new super-compressed format for 3D Gaussian Splatting.

Goodbye SOGS, hello SOG! 👋 PlayCanvas open sources Spatially Ordered Gaussians - a new super-compressed format for 3D Gaussian Splatting.

PlayCanvas

48,260 Aufrufe • vor 9 Monaten

3D Gaussian Splatting in WebGPU is ready

3D Gaussian Splatting in WebGPU is ready

el.cine

28,817 Aufrufe • vor 5 Monaten

Photogrammetry × 3D Gaussian Splatting × VRC Light Volumes！！

Photogrammetry × 3D Gaussian Splatting × VRC Light Volumes！！

とこよし

76,297 Aufrufe • vor 11 Monaten

Gaussian Shell Maps are a new neural scene representation that connects fields and 3D Gaussians. This representation unlocks the full potential of 3D Gaussian splatting for generative AI applications, such as 3D avatar generation. 1/2

Gordon Wetzstein

52,449 Aufrufe • vor 2 Jahren

a two minute explanation of 3D Gaussian Splatting

a two minute explanation of 3D Gaussian Splatting

dylan

254,807 Aufrufe • vor 2 Jahren

Fisheye Gaussian Splatting in PlayCanvas with 180° FOV in the browser. Step inside the scene… this is immersion on another level 🔥

Fisheye Gaussian Splatting in PlayCanvas with 180° FOV in the browser. Step inside the scene… this is immersion on another level 🔥

Martin Valigursky

10,715 Aufrufe • vor 2 Monaten

AnySplat Feed-forward 3D Gaussian Splatting from Unconstrained Views

AnySplat Feed-forward 3D Gaussian Splatting from Unconstrained Views

AK

18,703 Aufrufe • vor 1 Jahr

Is 3D Gaussian Splatting now the best medium for product visualization? 🤔 #playcanvas #supersplat

Is 3D Gaussian Splatting now the best medium for product visualization? 🤔 #playcanvas #supersplat

Will Eastcott

38,001 Aufrufe • vor 1 Jahr

GauS-SLAM: Dense RGB-D SLAM with Gaussian Surfels • We propose a 2D Gaussian-based incremental reconstruction strategy and a Surface-aware Depth Rendering mechanism. This approach effectively mitigates geometry distortions and improves tracking accuracy. • Our dense SLAM system features a front-end/back-end architecture and incorporates a local map design, ensuring tracking accuracy and efficiency. • We conduct extensive experiments demonstrating the superiority of our approach in both tracking accuracy and reconstruction quality compared to SOTA methods.

GauS-SLAM: Dense RGB-D SLAM with Gaussian Surfels • We propose a 2D Gaussian-based incremental reconstruction strategy and a Surface-aware Depth Rendering mechanism. This approach effectively mitigates geometry distortions and improves tracking accuracy. • Our dense SLAM system features a front-end/back-end architecture and incorporates a local map design, ensuring tracking accuracy and efficiency. • We conduct extensive experiments demonstrating the superiority of our approach in both tracking accuracy and reconstruction quality compared to SOTA methods.

MrNeRF

11,004 Aufrufe • vor 1 Jahr