Gaussian splatting
PulseAugur coverage of Gaussian splatting — every cluster mentioning Gaussian splatting across labs, papers, and developer communities, ranked by signal.
- instance of Gotit.pub 90%
- instance of CatalyzeX 90%
- used by Gotit.pub 90%
- uses Simultaneous localization and mapping 90%
- used by alphaXiv 70%
- instance of alphaXiv 70%
- instance of ScienceCast 70%
- used by ScienceCast 70%
- used by CatalyzeX 70%
- used by CORE Recommender 70%
- developed Simultaneous localization and mapping 70%
- competes with Nerf 60%
6 day(s) with sentiment data
-
SVRecon framework enables high-resolution neural surface reconstruction
Researchers have introduced Sparse Volumetric Reconstruction (SVRecon), a novel framework for generalizable neural surface reconstruction. This method addresses the memory limitations of previous approaches by employing…
-
New HLC-GS method improves satellite imagery DSM reconstruction
Researchers have developed HLC-GS, a novel method for reconstructing Digital Surface Models (DSMs) from optical satellite imagery using 3D Gaussian Splatting. This new approach addresses issues of height-layer mixing an…
-
New benchmarks and methods advance AI's 3D scene understanding capabilities · 3 sources tracked
Three new research papers introduce novel benchmarks and methods for improving 3D scene understanding in AI. SceneBench offers a hierarchical dataset with detailed annotations for evaluating vision-language models on sp…
-
ECCV 2026 highlights AI's shift to predicting and acting in the world · 1 source tracked
The ECCV 2026 conference is set to feature a significant surge in research on World Models and Embodied AI, with 13 workshops dedicated to these topics, reflecting a shift from understanding the world to predicting and …
-
FrameTwin framework uses Gaussian alignment for adaptive 3D printing
Researchers have developed FrameTwin, a novel framework for adaptive wireframe 3D printing. This system utilizes sparse-view images and curve-anchored Gaussian kernels to capture and model the deformation of thin wirefr…
-
New SLAM system integrates semantic understanding for dynamic environments
Researchers have developed RoSe-SLAM, a novel Simultaneous Localization and Mapping (SLAM) system designed to overcome the limitations of traditional methods in dynamic environments. By integrating semantic understandin…
-
New AI methods enhance 3D scene editing with multi-view consistency and LLM integration · 6 sources tracked
Researchers are developing new methods to improve the consistency and accuracy of 3D scene editing using AI. One approach, "3D-Consistent Multi-View Editing by Correspondence Guidance," focuses on ensuring that correspo…
-
NeRF method leads in 3D reconstruction for immersive education · 2 sources tracked
A recent study compared four 3D reconstruction methods—photogrammetry, NeRF-based, Gaussian splatting, and LiDAR—for creating realistic holographic representations of laboratory objects for educational use. The NeRF-bas…
-
New PAGS framework improves photoacoustic tomography with speed-of-sound adaptation
Researchers have developed PAGS, a novel framework for photoacoustic computed tomography (PACT) that addresses the challenge of unknown speed-of-sound heterogeneity. This method uses speed-of-sound-adaptive Gaussian spl…
-
New Gaussian Splatting Representation Integrates Physics into Neural Operators
Researchers have developed a novel approach to integrate physical laws into neural operator models for solving partial differential equations (PDEs). This method uses a feed-forward Gaussian splatting representation to …
-
FixAnything model refines 3D rendering artifacts using video generative priors
Researchers have developed FixAnything, a novel model designed to refine 3D renderings and fix artifacts across various representations like Gaussian Splatting and NeRF. The model repurposes a pre-trained video generati…
-
3D Gaussian Splatting research advances copyright, compression, and dynamic scene modeling · 10 sources tracked
Recent research is advancing 3D Gaussian Splatting (3DGS) through various innovations, including methods for copyright protection with X-SG$^2$S, improved reconstruction quality with DualDiff3D, and a theoretical framew…
-
New methods enhance VLA policy efficiency and robustness in robotics · 4 sources tracked
Researchers have developed new methods to improve the efficiency and robustness of vision-language-action (VLA) policies in robotics. One approach, EXIMO, uses a vision-language model (VLM) as a planner to break down co…
-
New research tackles 3D reconstruction from sparse views using novel methods · 2 sources tracked
Two new research papers, "Point-Based 3D Reconstruction from Sparse Views under Known Illumination" and "GeoRect4D: Geometry-Compatible Generative Rectification for Dynamic Sparse-View 3D Reconstruction," present novel …
-
LaGSplat framework infers physics-governed simulations from video
Researchers have developed LaGSplat, a novel framework that infers physics-governed interactive simulations from monocular video. This system allows users to apply external forces to filmed objects, enabling real-time r…
-
MotionGS-SLAM tackles motion blur in robotics with event cameras
Researchers have developed MotionGS-SLAM, a novel system for Simultaneous Localization and Mapping (SLAM) that effectively handles motion blur by modeling blur formation within its rendering pipeline. Unlike traditional…
-
New splat-based method reduces metal artifacts in CT scans
Researchers have developed a new method for reducing artifacts in cone-beam computed tomography (CBCT) scans, particularly those caused by metal implants. This novel approach utilizes a physics-inspired, self-calibratin…
-
SplitGaussian framework improves dynamic 3D scene reconstruction
Researchers have introduced SplitGaussian, a new framework designed to improve the reconstruction of dynamic 3D scenes from monocular video. Unlike previous methods that combine static and dynamic scene elements, SplitG…
-
XClipGS method enhances medical volume rendering with exact clipping
Researchers have developed XClipGS, a novel method for exact half-space clipping in medical volume Gaussian splatting. This technique allows for interactive rendering of volumetric medical scans by precisely clipping an…
-
TRACE method optimizes active scene reconstruction trajectories
Researchers have developed TRACE, a novel approach to active scene reconstruction that optimizes sensor trajectories for better information gathering. Unlike previous greedy methods that select the next best view in iso…