C4G introduces compact timestamp-conditioned Gaussian query tokens that aggregate full temporal context to decode 3D Gaussians with timestamp-modulated positions for feed-forward 4D reconstruction from monocular video, plus a diffusion-based rendering module and extension to 4D feature fields.
hub
V olsplat: Rethinking feed-forward 3d gaussian splatting with voxel-aligned prediction
19 Pith papers cite this work. Polarity classification is still indexing.
19
Pith papers citing it
abstract
Feed-forward 3D Gaussian Splatting (3DGS) has emerged as a highly effective solution for novel view synthesis. Existing methods predominantly rely on a \emph{pixel-aligned} Gaussian prediction paradigm, where each 2D pixel is mapped to a 3D Gaussian. We rethink this widely adopted formulation and identify several inherent limitations: it renders the reconstructed 3D models heavily dependent on the number of input views, leads to view-biased density distributions, and introduces alignment errors, particularly when source views contain occlusions or low texture. To address these challenges, we introduce VolSplat, a new multi-view feed-forward paradigm that replaces pixel alignment with voxel-aligned Gaussians. By directly predicting Gaussians from a predicted 3D voxel grid, it overcomes pixel alignment's reliance on error-prone 2D feature matching, ensuring robust multi-view consistency. Furthermore, it enables adaptive control over density based on 3D scene complexity, yielding more faithful Gaussians, improved geometric consistency, and enhanced novel-view rendering quality. Experiments on widely used benchmarks demonstrate that VolSplat achieves state-of-the-art performance, while producing more plausible and view-consistent results. The video results, code and trained models are available on our project page: https://lhmd.top/volsplat.
hub tools
citation-role summary
background 3
citation-polarity summary
fields
cs.CV 19verdicts
UNVERDICTED 19roles
background 3polarities
background 3representative citing papers
A meta-learned optimizer for 3DGS that extends the optimization horizon via checkpoint buffers and latent gradient-scale encoding, delivering better early novel-view quality and long-term stability with zero-shot generalization.
AdaptSplat adds a Frequency-Preserving Adapter to vision foundation models to boost high-frequency fidelity and cross-domain performance in feed-forward 3D Gaussian Splatting.
SplatWeaver uses cardinality Gaussian experts and pixel-level routing to dynamically allocate varying numbers of Gaussian primitives for generalizable novel view synthesis.
Mix3R mixes feed-forward reconstruction and generative 3D priors via Mixture-of-Transformers and overlap-based attention bias to achieve better-aligned 3D shapes and more accurate poses than either approach alone.
GSCompleter completes 3DGS scenes from sparse viewpoints using a generate-then-register workflow with stereo-anchor view selection and ray-constrained registration to achieve metric-aware results and SOTA performance on benchmarks.
GlobalSplat achieves competitive novel-view synthesis on RealEstate10K and ACID using only 16K Gaussians via global scene tokens and coarse-to-fine training, with a 4MB footprint and under 78ms inference.
AnchorSplat uses anchor-aligned 3D Gaussians guided by geometric priors for feed-forward scene reconstruction, achieving SOTA novel view synthesis on ScanNet++ with fewer primitives and better view consistency.
Free-Range Gaussians uses flow matching over Gaussian parameters to predict non-grid-aligned 3D Gaussians from multi-view images, enabling synthesis of plausible content in unobserved regions with fewer primitives than grid-aligned methods.
COVScene is a pose-free framework that lifts semantic Gaussians into a volumetric occupancy field during training to jointly support novel view synthesis, open-vocabulary segmentation, and semantic occupancy prediction.
HiReFF presents a feed-forward framework for 2K human video reconstruction from uncalibrated sparse-view videos via scale-synchronized calibration, Gaussian masking, and high-resolution side-tuning.
Mirage stores and queries 3D scene information in diffusion latent space via depth-guided lifting and warping, yielding 10.57× faster generation and 55× smaller memory than explicit RGB point-cloud baselines while reaching SOTA on WorldScore.
TriSplat predicts oriented triangle primitives from images in one forward pass to produce simulation-ready 3D meshes with competitive rendering quality.
Layer analysis of DINOv3 shows non-uniform 3D geometric knowledge concentrated in deeper layers, enabling a last-layer-centric recombination module that improves monocular depth estimation accuracy to state-of-the-art levels.
The paper proposes a problem-driven taxonomy for feed-forward 3D scene modeling that groups methods by five core challenges: feature enhancement, geometry awareness, model efficiency, augmentation strategies, and temporal-aware modeling.
C3G creates compact 3D Gaussian representations with 2K points by guiding placement via learnable tokens that aggregate multi-view features through attention, yielding better efficiency and performance than dense methods.
PhysRAG curates 7K videos from WISA-80K, builds a physical video database, and injects knowledge via learnable queries into a diffusion model to reach SOTA visual quality and physical compliance on PhyGenBench and VBench.
VG²GT regresses Gaussian primitive parameters from multi-scale voxel features of a frozen VFM and uses stochastic solid volume rendering for depth supervision to produce geometrically accurate reconstructions that outperform prior methods on DTU, Replica, TAT, and ScanNet.
UniMesh unifies 3D mesh generation and understanding in one model via a Mesh Head interface, Chain of Mesh iterative editing, and an Actor-Evaluator self-reflection loop.
citing papers explorer
-
Learning Global Motion with Compact Gaussians for Feed-Forward 4D Reconstruction
C4G introduces compact timestamp-conditioned Gaussian query tokens that aggregate full temporal context to decode 3D Gaussians with timestamp-modulated positions for feed-forward 4D reconstruction from monocular video, plus a diffusion-based rendering module and extension to 4D feature fields.
-
Learn2Splat: Extending the Horizon of Learned 3DGS Optimization
A meta-learned optimizer for 3DGS that extends the optimization horizon via checkpoint buffers and latent gradient-scale encoding, delivering better early novel-view quality and long-term stability with zero-shot generalization.
-
AdaptSplat: Adapting Vision Foundation Models for Feed-Forward 3D Gaussian Splatting
AdaptSplat adds a Frequency-Preserving Adapter to vision foundation models to boost high-frequency fidelity and cross-domain performance in feed-forward 3D Gaussian Splatting.
-
SplatWeaver: Learning to Allocate Gaussian Primitives for Generalizable Novel View Synthesis
SplatWeaver uses cardinality Gaussian experts and pixel-level routing to dynamically allocate varying numbers of Gaussian primitives for generalizable novel view synthesis.
-
Mix3R: Mixing Feed-forward Reconstruction and Generative 3D Priors for Joint Multi-view Aligned 3D Reconstruction and Pose Estimation
Mix3R mixes feed-forward reconstruction and generative 3D priors via Mixture-of-Transformers and overlap-based attention bias to achieve better-aligned 3D shapes and more accurate poses than either approach alone.
-
GSCompleter: A Distillation-Free Plugin for Metric-Aware 3D Gaussian Splatting Completion in Seconds
GSCompleter completes 3DGS scenes from sparse viewpoints using a generate-then-register workflow with stereo-anchor view selection and ray-constrained registration to achieve metric-aware results and SOTA performance on benchmarks.
-
GlobalSplat: Efficient Feed-Forward 3D Gaussian Splatting via Global Scene Tokens
GlobalSplat achieves competitive novel-view synthesis on RealEstate10K and ACID using only 16K Gaussians via global scene tokens and coarse-to-fine training, with a 4MB footprint and under 78ms inference.
-
AnchorSplat: Feed-Forward 3D Gaussian Splatting with 3D Geometric Priors
AnchorSplat uses anchor-aligned 3D Gaussians guided by geometric priors for feed-forward scene reconstruction, achieving SOTA novel view synthesis on ScanNet++ with fewer primitives and better view consistency.
-
Free-Range Gaussians: Non-Grid-Aligned Generative 3D Gaussian Reconstruction
Free-Range Gaussians uses flow matching over Gaussian parameters to predict non-grid-aligned 3D Gaussians from multi-view images, enabling synthesis of plausible content in unobserved regions with fewer primitives than grid-aligned methods.
-
Bridging 3D Gaussians and Semantic Occupancy for Comprehensive Open-Vocabulary Scene Understanding from Unposed Images
COVScene is a pose-free framework that lifts semantic Gaussians into a volumetric occupancy field during training to jointly support novel view synthesis, open-vocabulary segmentation, and semantic occupancy prediction.
-
HiReFF: High-Resolution Feedforward Human Reconstruction from Uncalibrated Sparse-View Video
HiReFF presents a feed-forward framework for 2K human video reconstruction from uncalibrated sparse-view videos via scale-synchronized calibration, Gaussian masking, and high-resolution side-tuning.
-
Latent Spatial Memory for Video World Models
Mirage stores and queries 3D scene information in diffusion latent space via depth-guided lifting and warping, yielding 10.57× faster generation and 55× smaller memory than explicit RGB point-cloud baselines while reaching SOTA on WorldScore.
-
TriSplat: Simulation-Ready Feed-Forward 3D Scene Reconstruction
TriSplat predicts oriented triangle primitives from images in one forward pass to produce simulation-ready 3D meshes with competitive rendering quality.
-
Last-Layer-Centric Feature Recombination: Unleashing 3D Geometric Knowledge in DINOv3 for Monocular Depth Estimation
Layer analysis of DINOv3 shows non-uniform 3D geometric knowledge concentrated in deeper layers, enabling a last-layer-centric recombination module that improves monocular depth estimation accuracy to state-of-the-art levels.
-
Feed-Forward 3D Scene Modeling: A Problem-Driven Perspective
The paper proposes a problem-driven taxonomy for feed-forward 3D scene modeling that groups methods by five core challenges: feature enhancement, geometry awareness, model efficiency, augmentation strategies, and temporal-aware modeling.
-
C3G: Learning Compact 3D Representations with 2K Gaussians
C3G creates compact 3D Gaussian representations with 2K points by guiding placement via learnable tokens that aggregate multi-view features through attention, yielding better efficiency and performance than dense methods.
-
PhysRAG: Enhancing Physics-Awareness in Video Generation via Retrieval-Augmented Generation
PhysRAG curates 7K videos from WISA-80K, builds a physical video database, and injects knowledge via learnable queries into a diffusion model to reach SOTA visual quality and physical compliance on PhyGenBench and VBench.
-
$\text{VG}^2$GT: Voxel-Gaussian Splatting Visual Geometry Grounded Transformer
VG²GT regresses Gaussian primitive parameters from multi-scale voxel features of a frozen VFM and uses stochastic solid volume rendering for depth supervision to produce geometrically accurate reconstructions that outperform prior methods on DTU, Replica, TAT, and ScanNet.
-
UniMesh: Unifying 3D Mesh Understanding and Generation
UniMesh unifies 3D mesh generation and understanding in one model via a Mesh Head interface, Chain of Mesh iterative editing, and an Actor-Evaluator self-reflection loop.