A method to construct propagation-consistent wireless environment digital twins from sparse CSI by creating a geometry-prior Bayesian channel map and calibrating a scene-level EM property field via differentiable ray tracing.
hub
3d gaussian splatting for real-time radiance field rendering
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
years
2026 10roles
background 1polarities
background 1representative citing papers
GaussLock embeds traps targeting position, scale, rotation, opacity, and color in 3D Gaussian models to degrade unauthorized fine-tunes while preserving authorized performance.
CARLA-GS is a modular pipeline that uses an LLM for semantic trajectory planning, CARLA for physics execution, and 3D Gaussian Splatting for photorealistic rendering to synthesize autonomous driving corner cases.
GeoGS-SLAM removes appearance parameters from 3D Gaussian Splatting for geometry-only dense monocular SLAM, achieving faster convergence and fewer primitives while introducing a coherent Sim(3) map update for loop closure.
GaussianMap learns adaptive Gaussian primitives on the BEV plane from multi-sensor data to produce vectorized HD maps, reporting state-of-the-art results on nuScenes and Argoverse 2.
SemDynReg constructs per-object ID maps from SAM and image features to regularize position, scale, and rotation of top-k Gaussians per object in dynamic 3DGS.
Ground4D reconstructs dynamic 4D scenes from monocular video by initializing dynamic Gaussians from VGGT geometry and refining them with multi-view depth consistency at observed and virtual viewpoints.
A hybrid structural latent points representation is learned by inserting a point-wise latent VAE into a point-cloud autoencoder and regularizing toward a Gaussian prior, paired with a lightweight 3DGS rendering pipeline, yielding gains on RLBench and ManiSkill2 benchmarks.
A surface extraction framework reduces the navigation state space by over 80% while achieving 100% planning success and sub-millisecond A* searches in Matterport3D and PCT scenes.
ReefMapGS closes the loop between multimodal SLAM and 3D Gaussian Splatting to deliver COLMAP-free incremental 3D reconstruction and improved AUV trajectory estimates for underwater reef surveys up to 700 m.
citing papers explorer
-
Propagation-Consistent Wireless Environment Digital Twin Construction Under Sparse Measurements
A method to construct propagation-consistent wireless environment digital twins from sparse CSI by creating a geometry-prior Bayesian channel map and calibrating a scene-level EM property field via differentiable ray tracing.
-
Immunizing 3D Gaussian Generative Models Against Unauthorized Fine-Tuning via Attribute-Space Traps
GaussLock embeds traps targeting position, scale, rotation, opacity, and color in 3D Gaussian models to degrade unauthorized fine-tunes while preserving authorized performance.
-
CARLA-GS: Decoupling Representation, Reasoning, and Physics Simulation for Autonomous Driving Corner-Case Synthesis
CARLA-GS is a modular pipeline that uses an LLM for semantic trajectory planning, CARLA for physics execution, and 3D Gaussian Splatting for photorealistic rendering to synthesize autonomous driving corner cases.
-
GeoGS-SLAM: Geometry-Only Gaussian Splatting for Dense Monocular SLAM
GeoGS-SLAM removes appearance parameters from 3D Gaussian Splatting for geometry-only dense monocular SLAM, achieving faster convergence and fewer primitives while introducing a coherent Sim(3) map update for loop closure.
-
GaussianMap: Learning Gaussian Representation for Multi-Sensor Online HD Map Construction
GaussianMap learns adaptive Gaussian primitives on the BEV plane from multi-sensor data to produce vectorized HD maps, reporting state-of-the-art results on nuScenes and Argoverse 2.
-
SemDynReg: Semantics-Guided Deformation Regularization for Dynamic 3D Gaussian Splatting
SemDynReg constructs per-object ID maps from SAM and image features to regularize position, scale, and rotation of top-k Gaussians per object in dynamic 3DGS.
-
Ground4D: Consistency-Aware 4D Reconstruction from Monocular Video
Ground4D reconstructs dynamic 4D scenes from monocular video by initializing dynamic Gaussians from VGGT geometry and refining them with multi-view depth consistency at observed and virtual viewpoints.
-
Learning Structural Latent Points for Efficient Visual Representations in Robotic Manipulation
A hybrid structural latent points representation is learned by inserting a point-wise latent VAE into a point-cloud autoencoder and regularizing toward a Gaussian prior, paired with a lightweight 3DGS rendering pipeline, yielding gains on RLBench and ManiSkill2 benchmarks.
-
Beyond Geometry: Efficient Topologically-Grounded Navigation in Complex 3D Environments
A surface extraction framework reduces the navigation state space by over 80% while achieving 100% planning success and sub-millisecond A* searches in Matterport3D and PCT scenes.
-
ReefMapGS: Enabling Large-Scale Underwater Reconstruction by Closing the Loop Between Multimodal SLAM and Gaussian Splatting
ReefMapGS closes the loop between multimodal SLAM and 3D Gaussian Splatting to deliver COLMAP-free incremental 3D reconstruction and improved AUV trajectory estimates for underwater reef surveys up to 700 m.