Skip to content
Preprint

Fusion-Aware Direct 3D Gaussian Generation with Structured Patch Latent Flows

Aug 2026 · 0 citations · 40 references
Computer Science

TL;DR

A structure-aware rectified flow model with patch-position conditioning, global-local coupled velocity prediction, and density-aware velocity weighting is designed, enabling direct latent generation of class-conditioned 3DGS objects within seconds.

Abstract

Class-guided 3D object generation is important for intelligent content creation, virtual environments, and digital asset design. Although 3D Gaussian Splatting (3DGS) offers an explicit and render-efficient representation, directly generating 3D Gaussian objects is difficult because Gaussian primitives are unordered, variable-sized, locally dense, and highly sensitive to rendering. Existing 3DGS generation methods usually depend on multi-view synthesis, reconstruction, or lifted 2D priors, fusing information mainly from observed views rather than modeling the intrinsic structural distribution of 3D Gaussian objects. This paper proposes a fusion-aware hierarchical Gaussian patch representation for direct class-guided 3DGS generation with rectified flow. Irregular Gaussian sets are decomposed into canonical local patches and encoded as structured tokens. The resulting hierarchical latent space fuses global class semantics, patch-level geometry and appearance, spatial correspondence, and rendering-sensitive cues. On this basis, we design a structure-aware rectified flow model with patch-position conditioning, global-local coupled velocity prediction, and density-aware velocity weighting, enabling direct latent generation of class-conditioned 3DGS objects within seconds. A render-feedback fusion strategy further aligns latent flow learning with decoded multi-view rendering quality. Experiments show that the proposed method generates 3D Gaussian objects with more coherent geometry, sharper local details, and better multi-view consistency than baseline latent generative models. Ablation studies confirm the contributions of hierarchical information fusion, global-local coupling, density-aware supervision, and render-feedback learning while preserving practical sampling efficiency overall.

View source

Similar papers

Preprint Aug 2026

Visibility-Guided Structured Measure Flow for Class-Conditioned 3D Gaussian Generation

3D Gaussian Splatting (3DGS) has made real-time, high-fidelity 3D rendering practical, yet turning this explicit representation into a native generative space remains an open challenge. Directly generating 3DGS objects is difficult because Gaussian primitives are unordered, variable-sized, locally dense, and highly sen...

Yi-Zhao Wang · 0 citations
Preprint Aug 2026

WilLaGS: Latent-Conditional 3D Appearance Fields for Robust Gaussian Splatting In-the-Wild

A generative appearance model where a $\beta$-VAE learns a structured and continuous manifold of global appearance is introduced, Conditioned on the latent code, a 3D neural appearance field is constructed that generates dynamic Tri-Plane features to encode spatially-varying local illumination effects.

Yu Bai, Qian-Qiu Tan, Li-Long Chen et al. · 1 citation
Sep 2026

D3GS: Depth, DINO, and RGB Diffusion Co-Guided 3D Gaussian Splatting for Sparse-View Reconstruction

Novel view synthesis from sparse inputs remains challenging for 3D Gaussian Splatting (3DGS) due to ambiguous geometry, cross-view inconsistency, and missing details in under-constrained regions, resulting in degraded reconstruction and unstable rendering. To tackle these issues, we propose D$^{3}$GS, a Depth-DINO-Diff...

Yun-Qi Gao, Zhan-Feng Liao, Han-Zhang Tu et al. · 0 citations
Preprint Aug 2026

GaussianDS: Depth-supervised Semantic Gaussian Splatting for Scene Understanding

GaussianDS, a depth-supervised semantic 3DGS framework that treats semantic lifting as a supervision-alignment problem and jointly optimizes RGB appearance, rendered depth, and compact semantics from scratch, is proposed.

Yu-Fei Zhang, Chen-Lu Zhan, Hong-Wei Wang · 0 citations
Preprint Aug 2026

Compact Feed-Forward 3D Gaussians via Saliency-Guided Primitive Merging

A structure-aware merging pipeline that takes per-pixel primitives from any feed-forward method and consolidates them into a compact, content-adaptive Gaussian set while largely retaining visual quality at just $\frac{1}{20}^\text{th}$ of the Gaussians of a per-pixel method is presented.

Tim-Felix Faasch, Jochen Kall, C. Stachniss · 0 citations
Preprint Aug 2026

Luce: Relightable Gaussians for 3D Asset Generation

Luce is proposed, a 3D representation that unifies geometry and PBR materials within a voxelized multimodal Gaussian cloud, using dedicated Gaussian primitives for albedo, metallic-roughness, and surface normals to achieve state-of-the-art single-image-to-3D generation.

M. Singh, Michele Stoppa, Alvise Memo et al. · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.