# FlowR: Flowing from Sparse to Dense 3D Reconstructions

arXiv: https://arxiv.org/abs/2504.01647 · HTML: https://mlsanity.com/paper/2504.01647/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 61 of its 74 references that are in the map.
- First posted on arXiv 2 Apr 2025.

## Builds on (most cited first, 30 of 61)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [Scalable Diffusion Models with Transformers](https://mlsanity.com/paper/2212.09748/) (2022)
- [Flow Matching for Generative Modeling](https://mlsanity.com/paper/2210.02747/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Scaling Rectified Flow Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2403.03206/) (2024)
- [Flow Straight and Fast: Learning to Generate and Transfer Data with Rectified Flow](https://mlsanity.com/paper/2209.03003/) (2022)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [DreamFusion: Text-to-3D using 2D Diffusion](https://mlsanity.com/paper/2209.14988/) (2022)
- [FlashAttention-2: Faster Attention with Better Parallelism and Work Partitioning](https://mlsanity.com/paper/2307.08691/) (2023)
- [Stable Video Diffusion: Scaling Latent Video Diffusion Models to Large Datasets](https://mlsanity.com/paper/2311.15127/) (2023)
- [CogVideoX: Text-to-Video Diffusion Models with An Expert Transformer](https://mlsanity.com/paper/2408.06072/) (2024)
- [Mip-NeRF 360: Unbounded Anti-Aliased Neural Radiance Fields](https://mlsanity.com/paper/2111.12077/) (2021)
- [pixelNeRF: Neural Radiance Fields from One or Few Images](https://mlsanity.com/paper/2012.02190/) (2020)
- [Align your Latents: High-Resolution Video Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2304.08818/) (2023)
- [ProlificDreamer: High-Fidelity and Diverse Text-to-3D Generation with Variational Score Distillation](https://mlsanity.com/paper/2305.16213/) (2023)
- [Building Normalizing Flows with Stochastic Interpolants](https://mlsanity.com/paper/2209.15571/) (2022)
- [Grounding Image Matching in 3D with MASt3R](https://mlsanity.com/paper/2406.09756/) (2024)
- [LRM: Large Reconstruction Model for Single Image to 3D](https://mlsanity.com/paper/2311.04400/) (2023)
- [ScanNet++: A High-Fidelity Dataset of 3D Indoor Scenes](https://mlsanity.com/paper/2308.11417/) (2023)
- [MVSNeRF: Fast Generalizable Radiance Field Reconstruction from Multi-View Stereo](https://mlsanity.com/paper/2103.15595/) (2021)
- [Extending Context Window of Large Language Models via Positional Interpolation](https://mlsanity.com/paper/2306.15595/) (2023)
- [LGM: Large Multi-View Gaussian Model for High-Resolution 3D Content Creation](https://mlsanity.com/paper/2402.05054/) (2024)
- [pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction](https://mlsanity.com/paper/2312.12337/) (2023)
- [Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion](https://mlsanity.com/paper/2407.01392/) (2024)
- [Zip-NeRF: Anti-Aliased Grid-Based Neural Radiance Fields](https://mlsanity.com/paper/2304.06706/) (2023)
- [Movie Gen: A Cast of Media Foundation Models](https://mlsanity.com/paper/2410.13720/) (2024)

## Built on by (most cited first, 28 of 28)

- [G4Splat: Geometry-Guided Gaussian Splatting with Generative Prior](https://mlsanity.com/paper/2510.12099/) (2025)
- [DiffusionHarmonizer: Bridging Neural Reconstruction and Photorealistic Simulation with Online Diffusion Enhancer](https://mlsanity.com/paper/2602.24096/) (2026)
- [LagerNVS: Latent Geometry for Fully Neural Real-time Novel View Synthesis](https://mlsanity.com/paper/2603.20176/) (2026)
- [World Reconstruction From Inconsistent Views](https://mlsanity.com/paper/2603.16736/) (2026)
- [GaussFusion: Improving 3D Reconstruction in the Wild with A Geometry-Informed Video Generator](https://mlsanity.com/paper/2603.25053/) (2026)
- [LSD-3D: Large-Scale 3D Driving Scene Generation with Geometry Grounding](https://mlsanity.com/paper/2508.19204/) (2025)
- [From Orbit to Ground: Generative City Photogrammetry from Extreme Off-Nadir Satellite Images](https://mlsanity.com/paper/2512.07527/) (2025)
- [Sparse-View 3D Reconstruction: Recent Advances and Open Challenges](https://mlsanity.com/paper/2507.16406/) (2025)
- [SimULi: Real-Time LiDAR and Camera Simulation with Unscented Transforms](https://mlsanity.com/paper/2510.12901/) (2025)
- [GIFSplat: Generative Prior-Guided Iterative Feed-Forward 3D Gaussian Splatting from Sparse Views](https://mlsanity.com/paper/2602.22571/) (2026)
- [Pantheon360: Taming Digital Twin Generation via 3D-Aware 360deg Video Diffusion](https://mlsanity.com/paper/2605.25449/) (2026)
- [GaussianLens: Localized High-Resolution Reconstruction via On-Demand Gaussian Densification](https://mlsanity.com/paper/2509.25603/) (2025)
- [Confidence-Based Mesh Extraction from 3D Gaussians](https://mlsanity.com/paper/2603.24725/) (2026)
- [PhaseFlow4D: Physically Constrained 4D Beam Reconstruction via Feedback-Guided Latent Diffusion](https://mlsanity.com/paper/2604.03885/) (2026)
- [GeodesicNVS: Probability Density Geodesic Flow Matching for Novel View Synthesis](https://mlsanity.com/paper/2603.01010/) (2026)
- [GSCompleter: A Distillation-Free Plugin for Metric-Aware 3D Gaussian Splatting Completion in Seconds](https://mlsanity.com/paper/2604.20155/) (2026)
- [From Knowledge to Action: Outcomes of the 2025 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry](https://mlsanity.com/paper/2605.03205/) (2026)
- [HAD: Hallucination-Aware Diffusion Priors for 3D Reconstruction](https://mlsanity.com/paper/2605.16873/) (2026)
- [Latent Riemannian Flow Matching for Geometry-Grounded 3D Foundation Models](https://mlsanity.com/paper/2607.19120/) (2026)
- [OGGSplat: Open Gaussian Growing for Generalizable Reconstruction with Expanded Field-of-View](https://mlsanity.com/paper/2506.05204/) (2025)
- [Leveling3D: Leveling Up 3D Reconstruction with Feed-Forward 3D Gaussian Splatting and Geometry-Aware Generation](https://mlsanity.com/paper/2603.16211/) (2026)
- [SatSplatDiff: Geometry-preserving generative refinement for high-fidelity satellite Gaussian Splatting](https://mlsanity.com/paper/2606.27223/) (2026)
- [World from Motion: Generative Dynamic Gaussian Reconstruction from Monocular Video](https://mlsanity.com/paper/2607.01202/) (2026)
- [MACRO: Training-free Multi-plane Attention for Closeup Render Optimization](https://mlsanity.com/paper/2607.03875/) (2026)
- [LightCrafter: PBR-Conditioned Video Diffusion Refinement for Controllable and Consistent Relighting](https://mlsanity.com/paper/2607.08016/) (2026)
- [GenRec: Knowing Where to Reconstruct and Where to Generate](https://mlsanity.com/paper/2608.17832/) (2026)
- [FixAnything: 3D-Consistent Rendering Refinement via Video Generative Priors](https://mlsanity.com/paper/2608.23549/) (2026)
- [Rethinking 3D Noise: Learning 3D-Aware Video Priors via Optimization-Free Morphological Perturbations](https://mlsanity.com/paper/2609.03657/) (2026)

## Abstract (arXiv)

3D Gaussian splatting enables high-quality novel view synthesis (NVS) at real-time frame rates. However, its quality drops sharply as we depart from the training views. Thus, dense captures are needed to match the high-quality expectations of applications like Virtual Reality (VR). However, such dense captures are very laborious and expensive to obtain. Existing works have explored using 2D generative models to alleviate this requirement by distillation or generating additional training views. These models typically rely on a noise-to-data generative process conditioned only on a handful of reference input views, leading to hallucinations, inconsistent generation results, and subsequent reconstruction artifacts. Instead, we propose a multi-view, flow matching model that learns a flow to directly connect novel view renderings from possibly sparse reconstructions to renderings that we expect from dense reconstructions. This enables augmenting scene captures with consistent, generated views to improve reconstruction quality. Our model is trained on a novel dataset of 3.6M image pairs and can process up to 45 views at 540x960 resolution (91K tokens) on one H100 GPU in a single forward pass. Our pipeline consistently improves NVS in sparse- and dense-view scenarios, leading to higher-quality reconstructions than prior works across multiple, widely-used NVS benchmarks.