PulseAugur
LIVE 21:00:12
research · [7 sources] ·

New 3D Gaussian Splatting Methods Enhance Scale, Object Awareness, and Editing

By PulseAugur Editorial · Summary by gemini-2.5-flash-lite from 7 sources

Researchers have developed several advancements in 3D Gaussian Splatting (3DGS) technology. TideGS enables training with over a billion primitives on a single GPU by managing parameters across SSD, CPU, and GPU. OP2GS introduces object-aware primitives with dual opacity for better scene understanding and editing. AnyCity addresses challenges in reconstructing large-scale urban scenes from sparse aerial views by predicting observation-supported geometry and using a diffusion prior. Additionally, 3D Skew Gaussian Splatting (3DSGS) enhances structural fidelity and compactness with asymmetric Gaussian primitives, while GaussianZoom offers progressive zoom-in capabilities with geometric and semantic guidance. Finally, a new framework leverages SAM-HQ and prior-guided label reassignment for robust segmentation in editable 3DGS. AI

Summary written by gemini-2.5-flash-lite from 7 sources. How we write summaries →

IMPACT These advancements push the boundaries of 3D scene reconstruction, enabling larger scales, better object understanding, and more sophisticated editing capabilities.

RANK_REASON Multiple research papers introduce novel techniques and improvements to 3D Gaussian Splatting.

Read on arXiv cs.CV →

New 3D Gaussian Splatting Methods Enhance Scale, Object Awareness, and Editing

COVERAGE [7]

  1. arXiv cs.CV TIER_1 · Tao Guan ·

    AIR: Amortized Image Reconstruction Framework for Self-Supervised Feed-Forward 2D Gaussian Splatting

    2D Gaussian splatting provides an efficient explicit representation for image reconstruction, but existing methods still require costly per-image iterative optimization or rely on handcrafted priors for primitive allocation. We present AIR, a self-supervised feed-forward framewor…

  2. arXiv cs.CV TIER_1 · Chaojian Li ·

    TideGS: Scalable Training of Over One Billion 3D Gaussian Splatting Primitives via Out-of-Core Optimization

    Training 3D Gaussian Splatting (3DGS) at billion-primitive scale is fundamentally memory-bound: each Gaussian primitive carries a large attribute vector, and the aggregate parameter table quickly exceeds GPU capacity, limiting prior systems to tens of millions of Gaussians on com…

  3. arXiv cs.CV TIER_1 · Janne Heikkilä ·

    OP2GS: Object-Aware 3D Gaussian Splatting with Dual-Opacity Primitives

    3D Gaussian Splatting (3DGS) provides an explicit and efficient scene representation, but its primitives lack inherent object-level identity, hindering downstream tasks such as open-vocabulary scene understanding. Existing methods typically address this by either distilling high-…

  4. arXiv cs.CV TIER_1 · Wufan Zhao ·

    Feed-Forward Gaussian Splatting from Sparse Aerial Views

    Reconstructing large-scale urban scenes from sparse aerial views is a crucial yet challenging task. Due to biased top-down and shallow-oblique camera poses, sparse aerial captures exhibit strong evidence imbalance: roofs and open regions are repeatedly observed, while facades, di…

  5. arXiv cs.CV TIER_1 · Hao Wang ·

    3D Skew Gaussian Splatting with Any Camera Trajectory Visualization Engine

    While 3D Gaussian Splatting (3DGS) has revolutionized real-time photorealistic view synthesis, its fundamental reliance on symmetric Gaussian distributions introduces visual artifacts that hinder accurate spatial data exploration. Specifically, symmetric kernels struggle to captu…

  6. arXiv cs.CV TIER_1 · Zhaopeng Cui ·

    GaussianZoom: Progressive Zoom-in Generative 3D Gaussian Splatting with Geometric and Semantic Guidance

    We introduce GaussianZoom, a generative zoom-in 3D reconstruction system with an iterative progressive framework that combines geometry-consistent scene modeling and multi-scale semantic reasoning to enable high-fidelity extreme zoom-in rendering from low-resolution inputs. To ac…

  7. arXiv cs.CV TIER_1 · Jean-Yves Guillemaut ·

    Robust Prior-Guided Segmentation for Editable 3D Gaussian Splatting

    3D Gaussian Splatting (3D-GS) enables real-time 3D scene reconstruction but lacks robust segmentation for editing tasks such as object removal, extraction, and recoloring. Existing approaches that lift 2D segmentations to the 3D domain suffer from view inconsistencies and coarse …

RELATED ENTITIES

RELATED TOPICS