3D Gaussian splatting
PulseAugur coverage of 3D Gaussian splatting — every cluster mentioning 3D Gaussian splatting across labs, papers, and developer communities, ranked by signal.
- competes with Neural Radiance Fields 70%
- used by lidar 70%
- used by ScienceCast 70%
- instance of replica 70%
- uses Simultaneous localization and mapping 70%
- used by alphaXiv 70%
- competes with Nerf 60%
- used by DagsHub 60%
- other Neural Radiance Fields 50%
- used by Nerf 50%
- affiliated with lidar 50%
- affiliated with Simultaneous localization and mapping 50%
- 2026-05-15 research_milestone Publication of a research paper introducing a new method for robust prior-guided segmentation for editable 3D Gaussian Splatting. source
20 day(s) with sentiment data
-
CustomX system enables unified character, action, and scene customization in video world models
Researchers have introduced CustomX, a novel system that unifies character, action, and scene customization within video world models. This advancement allows users to provide a 3D scene and a character, then use natura…
-
New dataset enables novel view synthesis for gastroendoscopy
Researchers have introduced the GastroNVS dataset, the first real-world dataset designed for novel view synthesis (NVS) in gastroendoscopy. This dataset aims to address the current lack of sufficient data for evaluating…
-
New SkyLume Dataset Tackles 3D Urban Reconstruction Under Varying Light
Researchers have introduced SkyLume, a large-scale aerial dataset designed to address challenges in 3D urban scene reconstruction under varying illumination conditions. The dataset comprises over 100,000 high-resolution…
-
Render-FM achieves real-time photorealistic CT scan rendering
Researchers have developed Render-FM, a novel feedforward model designed for real-time photorealistic volumetric rendering of CT scans. This model significantly speeds up the rendering process, reducing it from hours or…
-
TryOnCrafter framework enables camera-controllable video virtual try-on · 3 sources tracked
Researchers have introduced TryOnCrafter, a novel framework for camera-controllable video virtual try-on. This system moves beyond existing methods by decoupling human subjects from their environments using a renderable…
-
Pocket-SLAM tackles memory limits in 3DGS-SLAM for autonomous driving
Researchers have developed Pocket-SLAM, a novel method to improve the memory efficiency of 3D Gaussian Splatting for Simultaneous Localization and Mapping (SLAM). This approach addresses the issue of accumulating Gaussi…
-
New framework creates interactive digital twins from RGB-D videos
Researchers have developed ArtiTwinSplat, a new framework for creating interactive digital twins of objects from RGB-D videos. This method utilizes 3D Gaussian Splatting to maintain visual realism and an unsupervised pi…
-
MM-TRELLIS generates 3D vehicles using multi-modal sensor data · 2 sources tracked
Researchers have developed MM-TRELLIS, a novel method for generating realistic 3D vehicle models from autonomous driving data. This approach integrates multi-view images and LiDAR point clouds into native 3D generative …
-
3DCarGen generates detailed 3D car models from single images
Researchers have developed 3DCarGen, a new framework for generating high-quality 3D car models from single images. This method synthesizes multiple consistent views of a car and uses 3D Gaussian Splatting to create a co…
-
FLUX3D framework boosts 3D Gaussian Splatting generation fidelity · 3 sources tracked
Researchers have introduced FLUX3D, a novel framework designed to enhance the fidelity of 3D Gaussian Splatting (3DGS) generation from images. The system addresses two key bottlenecks in current methods: the representat…
-
MeGAS framework integrates thermomechanical dynamics into 3D Gaussian Splatting
Researchers have introduced MeGAS, a novel framework that integrates thermomechanical dynamics into 3D Gaussian Splatting (3DGS). This new approach augments 3DGS with temperature attributes and a heat advection-diffusio…
-
DrivingVoxels framework enhances dynamic scene reconstruction
Researchers have introduced DrivingVoxels, a new framework designed to improve the reconstruction of dynamic driving scenes. This method addresses limitations in existing approaches, such as the time-consuming nature of…
-
New AI methods tackle 3D and 4D reconstruction from single images and video
Two new research papers introduce novel approaches to 3D and 4D reconstruction from visual data. PRISM offers a feed-forward method for single-image 3D reconstruction, decomposing the task into a geometric prior and a l…
-
VAST AI open-sources TripoSplat for single-image 3D model generation
VAST AI has open-sourced TripoSplat, a technology capable of generating 3D Gaussian splatting (3DGS) models from a single image. This model is optimized for applications in gaming, XR, and simulation environments.
-
New PVFD failure mode identified in 3D Gaussian tomography
Researchers have identified a new failure mode in 3D Gaussian Splatting methods used for sparse-view computed tomography, termed Projection-Volume Fidelity Divergence (PVFD). This divergence occurs when rendered project…
-
New datasets and models advance sign language recognition and translation
Researchers have developed new methods for sign language recognition and translation. One approach uses a deep learning pipeline combining a VideoMAE video transformer for classifying sign gestures into English words an…
-
MMD-SLAM enhances Visual SLAM with structure-guided Gaussian mapping
Researchers have introduced MMD-SLAM, a novel Visual SLAM framework designed to enhance mapping quality and tracking robustness by incorporating structural information. This new system leverages the Atlanta World assump…
-
New benchmarks target precise UAV navigation and safety
Researchers have introduced two new benchmarks for evaluating unmanned aerial vehicle (UAV) vision-language navigation (VLN) capabilities. The first, UAV-VLN-FOV, focuses on the 'see-and-reach' aspect, assessing how pre…
-
New DF3DV-1K dataset advances distractor-free novel view synthesis research
Researchers have introduced DF3DV-1K, a large-scale dataset designed to advance research in distractor-free novel view synthesis. This dataset comprises over 1,000 scenes, each with both clean and cluttered image sets, …
-
FlowObject framework enhances 3D reconstruction with generative priors and observational fidelity
Researchers have introduced FlowObject, a novel framework designed to improve 3D object reconstruction from limited image captures. This method addresses the limitations of existing generative models, which can exhibit …