REVIEW 21 cited by
LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
3D content creation has achieved significant progress in terms of both quality and speed. Although current feed-forward models can produce 3D objects in seconds, their resolution is constrained by the intensive computation required during training. In this paper, we introduce Large Multi-View Gaussian Model (LGM), a novel framework designed to generate high-resolution 3D models from text prompts or single-view images. Our key insights are two-fold: 1) 3D Representation: We propose multi-view Gaussian features as an efficient yet powerful representation, which can then be fused together for differentiable rendering. 2) 3D Backbone: We present an asymmetric U-Net as a high-throughput backbone operating on multi-view images, which can be produced from text or single-view image input by leveraging multi-view diffusion models. Extensive experiments demonstrate the high fidelity and efficiency of our approach. Notably, we maintain the fast speed to generate 3D objects within 5 seconds while boosting the training resolution to 512, thereby achieving high-resolution 3D content generation.
Forward citations
Cited by 21 Pith papers
-
CORGI: Consistency-Aware 3D Dog Reconstruction from a Single Image in the Wild
CORGI reconstructs high-fidelity, animatable 3D dogs from a single in-the-wild image via canonical orbital generation, deformable 3DGS anchored to D-SMAL, and self-supervised generative repair, without 3D supervision.
-
MVGBench: Comprehensive Benchmark for Multi-view Generation Models
MVGBench evaluates multi-view generators through self-consistency of 3D reconstructions and uses this protocol to rank 12 models and build a better one.
-
GeoWorldAD: Geometry World Action Model for Autonomous Driving
Grounding an autonomous-driving action model in ego-aligned multi-scale 3D geometry and latent future-geometry tokens improves NAVSIM closed-loop PDMS/EPDMS over prior geometry- and world-model-based planners.
-
Instant NuRec: Feed-Forward 3D Gaussian Reconstruction for Driving Scene Simulation
A feed-forward model reconstructs a layered, simulation-ready 3D Gaussian world from multi-view driving video in ~1.5 s, with quality approaching per-scene optimized reconstruction.
-
Hallo4D: Multi-Modal Hallucination Mitigation for Consistent Spatio-Temporal Generation
Hallo4D uses vision-language models to detect and correct spatial and temporal mistakes in AI-generated 3D and 4D content, improving consistency without retraining the base generators.
-
PixGS: Pixel-Space Diffusion for Direct 3D Gaussian Splat Generation
PixGS is a single-stage pixel-space diffusion model that directly produces high-quality 3D Gaussian Splats from text or images in ~1s, outperforming multi-stage latent methods on standard benchmarks.
-
TInR: Exploring Tool-Internalized Reasoning in Large Language Models
TInR-U internalizes tool knowledge into LLMs via bidirectional alignment, supervised fine-tuning, and reinforcement learning, outperforming standard tool-integrated reasoning in both in-domain and out-of-domain evaluations.
-
Make Your MoVe: Make Your 3D Contents by Adapting Multi-View Diffusion Models to External Editing
A tuning-free dual-pipeline that injects original normal latents into an edited multi-view diffusion stream, preserving geometry during 2D-to-3D appearance editing.
-
Gaussian Variation Field Diffusion for High-fidelity Video-to-4D Synthesis
A video-to-4D model that encodes mesh animations into compact Gaussian variation latents and diffuses them conditioned on the video and a canonical Gaussian splat.
-
Masks make discriminative models great again!
Training a single-image 3D Gaussian splat model on visible regions only, using visibility masks from optimized per-scene splats, improves reconstruction quality in visible areas and stays competitive with full-scene models.
-
Nabla-R2D3: Effective and Efficient 3D Diffusion Alignment with 2D Rewards
Nabla-R2D3 aligns 3D-native diffusion models with human preferences by backpropagating multi-view 2D reward gradients through the denoising process, improving reward without destroying the pretrained 3D prior.
-
Render-FM: Feedforward Model for Real-time Photorealistic Volumetric Rendering
Render-FM predicts 6D Gaussian splatting parameters from a CT volume in a single feedforward pass, enabling real-time rendering with quality comparable to per-scan optimized methods.
-
A 3D Facial Reconstruction Evaluation Methodology: Comparing Smartphone Scans with Deep Learning Based Methods Using Geometry and Morphometry Criteria
A new benchmarking framework using geometric morphometrics shows smartphone 3D facial scans preserve shape better than deep learning reconstructions from 2D images.
-
Large Images are Gaussians: High-Quality Large Image Representation with Levels of 2D Gaussian Splatting
A two-level 2D Gaussian splatting method with direct covariance optimization fits large images with more Gaussian points and higher PSNR than prior Gaussian-based image representation.
-
Few-step Flow for 3D Generation via Marginal-Data Transport Distillation
MDT-dist distills a pretrained 3D flow model into a 1-2 step generator using velocity matching plus velocity distillation, cutting TRELLIS inference from 6.1s to 0.68s while approximately preserving generation quality.
-
Category-Aware 3D Object Composition with Disentangled Texture and Shape Multi-view Diffusion
C33D blends a 3D model with an object category by generating a fused front view, then using texture and shape multi-view diffusion plus adaptive inversion to reconstruct a novel, consistent 3D model.
-
SMPL Normal Map Is All You Need for Single-view Textured Human Reconstruction
By fine-tuning the LGM object reconstruction model with SMPL normal maps and a normal-map constraint, SEHR reconstructs full clothed 3D humans from one image in a single forward pass.
-
ShapeLLM-Omni: A Native Multimodal LLM for 3D Generation and Understanding
ShapeLLM-Omni unifies text, image, and 3D generation and understanding in one autoregressive LLM using discrete 3D tokens and a new 3D-Alpaca training dataset.
-
ConsistentDreamer: View-Consistent Meshes Through Balanced Multi-View Gaussian Optimization
Generating six reference views first, then combining closest-view SDS with pixel-level reconstruction and automatic uncertainty weights, improves cross-view consistency in single-image 3D mesh generation.
-
CA-World: Multi-Object Counterfactual Alignment for Efficient Interactive-Ready Reconstruction
The paper's stated CA-World counterfactual claim is absent from the body, which instead describes the SAM3D-Phys pipeline for multi-object interactive reconstruction and simulation.
-
SAT: Supervisor Regularization and Animation Augmentation for Two-process Monocular Texture 3D Human Reconstruction
A two-stage Gaussian-splatting framework with supervisor feature regularization and online animation augmentation improves monocular textured 3D human reconstruction on CustomHuman and THuman3.0.
Discussion (0). Continue with ORCID to comment.