3D Gaussian splatting
PulseAugur coverage of 3D Gaussian splatting — every cluster mentioning 3D Gaussian splatting across labs, papers, and developer communities, ranked by signal.
- instance of DagsHub 90%
- instance of alphaXiv 90%
- instance of ScienceCast 90%
- instance of Gotit.pub 90%
- used by alphaXiv 90%
- used by Gotit.pub 90%
- used by ScienceCast 90%
- instance of CatalyzeX 90%
- instance of Gaussian splatting 90%
- instance of DL3DV 90%
- uses Simultaneous localization and mapping 90%
- instance of 2DGS 90%
- 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
-
New framework BLaFA enables robots to grasp and use tools based on language instructions
Researchers have developed BLaFA, a novel framework designed to enable robots to perform dexterous grasping and tool use based on natural language instructions. This system translates open-vocabulary commands into speci…
-
InstantHand reconstructs animatable 3D hand avatars from single images
Researchers have developed InstantHand, a novel framework capable of reconstructing animatable 3D hand avatars from a single image. This method bypasses the need for multi-view capture or per-subject optimization, utili…
-
New AffineGS method enhances 3D Gaussian Splatting geometry
Researchers have developed AffineGS, a novel method to improve the geometric accuracy of 3D Gaussian Splatting (3DGS) for photorealistic novel view synthesis. The method addresses a fundamental flaw in existing depth-gu…
-
New WV-GS framework synthesizes wide views from multiple cameras for autonomous driving
Researchers have developed Wide-View Gaussian Splatting (WV-GS), a novel feed-forward framework for synthesizing a unified wide view from multiple overlapping perspective cameras, particularly for autonomous driving app…
-
MOMASPLAT framework enhances robot imitation learning with novel-view synthesis
Researchers have developed MOMASPLAT, a novel framework designed to enhance imitation learning for mobile manipulation tasks. This system addresses the challenge of robots performing tasks from viewpoints different from…
-
SplatAnything advances 3D Gaussian Splatting with novel normalization technique
Researchers have introduced SplatAnything, a novel method that enhances feed-forward 3D Gaussian Splatting (3DGS) by normalizing Gaussian spacing statistics. This approach addresses a key limitation in existing 3DGS tec…
-
New AEGIS framework protects 3D facial avatar privacy
Researchers have developed AEGIS, a novel framework designed to protect the privacy of 3D facial avatars. This system applies adversarial perturbations to the 3D Gaussian Splatting representation of avatars, effectively…
-
New WV-GS framework synthesizes wide views from multiple cameras for autonomous driving
Researchers have developed Wide-View Gaussian Splatting (WV-GS), a novel framework for synthesizing a unified wide view from multiple overlapping perspective cameras, a crucial capability for autonomous driving systems.…
-
New methods advance 3D Gaussian Splatting compression for dynamic scenes · 2 sources
Two new research papers introduce novel methods for compressing 3D Gaussian Splatting (3DGS) sequences, a technique used for rendering dynamic scenes. The first paper, GSCV++, focuses on lightweight compression using st…
-
DiTGaze system enhances few-shot gaze estimation using diffusion models
Researchers have developed DiTGaze, a novel system for few-shot gaze estimation that utilizes a two-stream 3D Gaussian Splatting backbone and a geometry-conditioned Diffusion Transformer. This approach reformulates the …
-
New AI frameworks enhance HDR video reconstruction and novel view synthesis
Researchers have developed HIVE-Diff, a diffusion framework for reconstructing high dynamic range (HDR) video from single low dynamic range (LDR) frames and event camera data. This method addresses limitations in existi…
-
LucidLyra framework enables queryable 3D scene reconstruction from video
Researchers have developed LucidLyra, a novel framework for 3D scene reconstruction that allows for semantic and instance querying directly from video latents. This approach distills knowledge from video diffusion model…
-
New ACD-GS framework optimizes 3D Gaussian Splatting for compact representations
Researchers have developed ACD-GS, a new framework for optimizing 3D Gaussian Splatting (3DGS) that addresses the issue of redundant Gaussian generation during training. Unlike previous methods that rely on screen-space…
-
New research advances 3D Gaussian Splatting for compression and real-world video
Recent research in 3D Gaussian Splatting (3DGS) focuses on improving compression, handling unsynchronized video, and enhancing feed-forward methods. GSCV++ offers lightweight compression for GS sequences by optimizing p…
-
Idol model generates 3D Gaussians from single images
Speridlabs has released Idol, a new model capable of generating up to 1,048,576 3D Gaussians from a single input image. This model can process renders, generated images, or photos, removing backgrounds and framing the o…
-
New method converts CT scans to real-time 3D Gaussian splats in minutes
Researchers have developed a new method called CTBakeGS that allows for the rapid conversion of computed tomography (CT) volume renderings into 3D Gaussian splatting (3DGS) models. This training-free approach significan…
-
GraftRay method integrates meshes into Gaussian Splatting scenes with accurate lighting
Researchers have developed GraftRay, a novel method for seamlessly integrating mesh objects into scenes reconstructed using Gaussian Splatting. This technique ensures physically accurate lighting and shadows between the…
-
ClusterSplat method improves 3D visual grounding in Gaussian Splatting
Researchers have developed ClusterSplat, a novel method for 3D visual grounding within 3D Gaussian Splatting (3DGS) environments. Unlike previous approaches that relied on heuristic filtering of relevance scores, Cluste…
-
New framework anonymizes sensitive data in 3D Gaussian Splatting
Researchers have developed PrinGS, a novel framework for preserving privacy in 3D Gaussian Splatting (3DGS) representations. This method allows for the anonymization of sensitive visual information, such as faces and te…
-
Hesai's Kosmo 3D scanner enters Hollywood, powering Pixar and embodied AI
Hesai Technology, a leading Chinese lidar company, has entered the film industry by providing its Kosmo 3D scanning solution to Pixar. This technology allows for the rapid and accurate capture of real-world environments…