# GFlow: Recovering 4D World from Monocular Video

arXiv: https://arxiv.org/abs/2405.18426 · HTML: https://mlsanity.com/paper/2405.18426/

## Key facts

- Cited by 46 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 42 references that are in the map.
- First posted on arXiv 28 May 2024.

## Builds on (most cited first, 27 of 27)

- [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)
- [DUSt3R: Geometric 3D Vision Made Easy](https://mlsanity.com/paper/2312.14132/) (2023)
- [D-NeRF: Neural Radiance Fields for Dynamic Scenes](https://mlsanity.com/paper/2011.13961/) (2020)
- [Nerfies: Deformable Neural Radiance Fields](https://mlsanity.com/paper/2011.12948/) (2020)
- [4D Gaussian Splatting for Real-Time Dynamic Scene Rendering](https://mlsanity.com/paper/2310.08528/) (2023)
- [Deformable 3D Gaussians for High-Fidelity Monocular Dynamic Scene Reconstruction](https://mlsanity.com/paper/2309.13101/) (2023)
- [K-Planes: Explicit Radiance Fields in Space, Time, and Appearance](https://mlsanity.com/paper/2301.10241/) (2023)
- [Dynamic 3D Gaussians: Tracking by Persistent Dynamic View Synthesis](https://mlsanity.com/paper/2308.09713/) (2023)
- [pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction](https://mlsanity.com/paper/2312.12337/) (2023)
- [BARF: Bundle-Adjusting Neural Radiance Fields](https://mlsanity.com/paper/2104.06405/) (2021)
- [HexPlane: A Fast Representation for Dynamic Scenes](https://mlsanity.com/paper/2301.09632/) (2023)
- [Real-time Photorealistic Dynamic Scene Representation and Rendering with 4D Gaussian Splatting](https://mlsanity.com/paper/2310.10642/) (2023)
- [GMFlow: Learning Optical Flow via Global Matching](https://mlsanity.com/paper/2111.13680/) (2021)
- [Unifying Flow, Stereo and Depth Estimation](https://mlsanity.com/paper/2211.05783/) (2022)
- [NoPe-NeRF: Optimising Neural Radiance Field with No Pose Prior](https://mlsanity.com/paper/2212.07388/) (2022)
- [DynIBaR: Neural Dynamic Image-Based Rendering](https://mlsanity.com/paper/2211.11082/) (2022)
- [Tensor4D : Efficient Neural 4D Decomposition for High-fidelity Dynamic Reconstruction and Rendering](https://mlsanity.com/paper/2211.11610/) (2022)
- [COLMAP-Free 3D Gaussian Splatting](https://mlsanity.com/paper/2312.07504/) (2023)
- [3DGStream: On-the-Fly Training of 3D Gaussians for Efficient Streaming of Photo-Realistic Free-Viewpoint Videos](https://mlsanity.com/paper/2403.01444/) (2024)
- [Layered Neural Atlases for Consistent Video Editing](https://mlsanity.com/paper/2109.11418/) (2021)
- [CoDeF: Content Deformation Fields for Temporally Consistent Video Processing](https://mlsanity.com/paper/2308.07926/) (2023)
- [4D Visualization of Dynamic Events from Unconstrained Multi-View Videos](https://mlsanity.com/paper/2005.13532/) (2020)
- [MonoNeRF: Learning a Generalizable Dynamic Radiance Field from Monocular Videos](https://mlsanity.com/paper/2212.13056/) (2022)
- [Im4D: High-Fidelity and Real-Time Novel View Synthesis for Dynamic Scenes](https://mlsanity.com/paper/2310.08585/) (2023)
- [Representing Volumetric Videos as Dynamic MLP Maps](https://mlsanity.com/paper/2304.06717/) (2023)
- [SiNeRF: Sinusoidal Neural Radiance Fields for Joint Pose Estimation and Scene Reconstruction](https://mlsanity.com/paper/2210.04553/) (2022)

## Built on by (most cited first, 30 of 46)

- [MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion](https://mlsanity.com/paper/2410.03825/) (2024)
- [Shape of Motion: 4D Reconstruction from a Single Video](https://mlsanity.com/paper/2407.13764/) (2024)
- [MegaSaM: Accurate, Fast, and Robust Structure and Motion from Casual Dynamic Videos](https://mlsanity.com/paper/2412.04463/) (2024)
- [FlashSplat: 2D to 3D Gaussian Splatting Segmentation Solved Optimally](https://mlsanity.com/paper/2409.08270/) (2024)
- [Feed-Forward Bullet-Time Reconstruction of Dynamic Scenes from Monocular Videos](https://mlsanity.com/paper/2412.03526/) (2024)
- [StereoCrafter: Diffusion-based Generation of Long and High-fidelity Stereoscopic 3D from Monocular Videos](https://mlsanity.com/paper/2409.07447/) (2024)
- [DAS3R: Dynamics-Aware Gaussian Splatting for Static Scene Reconstruction](https://mlsanity.com/paper/2412.19584/) (2024)
- [WristWorld: Generating Wrist-Views via 4D World Models for Robotic Manipulation](https://mlsanity.com/paper/2510.07313/) (2025)
- [HiMoR: Monocular Deformable Gaussian Reconstruction with Hierarchical Motion Representation](https://mlsanity.com/paper/2504.06210/) (2025)
- [4RC: 4D Reconstruction via Conditional Querying Anytime and Anywhere](https://mlsanity.com/paper/2602.10094/) (2026)
- [Test3R: Learning to Reconstruct 3D at Test Time](https://mlsanity.com/paper/2506.13750/) (2025)
- [iTACO: Interactable Digital Twins of Articulated Objects from Casually Captured RGBD Videos](https://mlsanity.com/paper/2506.08334/) (2025)
- [Poison-splat: Computation Cost Attack on 3D Gaussian Splatting](https://mlsanity.com/paper/2410.08190/) (2024)
- [Vista4D: Video Reshooting with 4D Point Clouds](https://mlsanity.com/paper/2604.21915/) (2026)
- [L4P: Towards Unified Low-Level 4D Vision Perception](https://mlsanity.com/paper/2502.13078/) (2025)
- [Efficient Gaussian Splatting for Monocular Dynamic Scene Rendering via Sparse Time-Variant Attribute Modeling](https://mlsanity.com/paper/2502.20378/) (2025)
- [Vivid4D: Improving 4D Reconstruction from Monocular Video by Video Inpainting](https://mlsanity.com/paper/2504.11092/) (2025)
- [TimeFormer: Capturing Temporal Relationships of Deformable 3D Gaussians for Robust Reconstruction](https://mlsanity.com/paper/2411.11941/) (2024)
- [C4D: 4D Made from 3D through Dual Correspondences](https://mlsanity.com/paper/2510.14960/) (2025)
- [RoDyGS: Robust Dynamic Gaussian Splatting for Casual Videos](https://mlsanity.com/paper/2412.03077/) (2024)
- [Seeing World Dynamics in a Nutshell](https://mlsanity.com/paper/2502.03465/) (2025)
- [PAD3R: Pose-Aware Dynamic 3D Reconstruction from Casual Videos](https://mlsanity.com/paper/2509.25183/) (2025)
- [WildSeg3D: Segment Any 3D Objects in the Wild from 2D Images](https://mlsanity.com/paper/2503.08407/) (2025)
- [Eye2Eye: A Simple Approach for Monocular-to-Stereo Video Synthesis](https://mlsanity.com/paper/2505.00135/) (2025)
- [Advances in Radiance Field for Dynamic Scene: From Neural Field to Gaussian Field](https://mlsanity.com/paper/2505.10049/) (2025)
- [Flow4DGS-SLAM: Optical Flow-Guided 4D Gaussian Splatting SLAM](https://mlsanity.com/paper/2604.22339/) (2026)
- [Mono4DGS-HDR: High Dynamic Range 4D Gaussian Splatting from Alternating-exposure Monocular Videos](https://mlsanity.com/paper/2510.18489/) (2025)
- [GP-4DGS: Probabilistic 4D Gaussian Splatting from Monocular Video via Variational Gaussian Processes](https://mlsanity.com/paper/2604.02915/) (2026)
- [Lift4D: Harmonizing Single-View 3D Estimation for 4D Reconstruction In-the-Wild](https://mlsanity.com/paper/2606.23688/) (2026)
- [Vista3D: Unravel the 3D Darkside of a Single Image](https://mlsanity.com/paper/2409.12193/) (2024)

## Abstract (arXiv)

Recovering 4D world from monocular video is a crucial yet challenging task. Conventional methods usually rely on the assumptions of multi-view videos, known camera parameters, or static scenes. In this paper, we relax all these constraints and tackle a highly ambitious but practical task: With only one monocular video without camera parameters, we aim to recover the dynamic 3D world alongside the camera poses. To solve this, we introduce GFlow, a new framework that utilizes only 2D priors (depth and optical flow) to lift a video to a 4D scene, as a flow of 3D Gaussians through space and time. GFlow starts by segmenting the video into still and moving parts, then alternates between optimizing camera poses and the dynamics of the 3D Gaussian points. This method ensures consistency among adjacent points and smooth transitions between frames. Since dynamic scenes always continually introduce new visual content, we present prior-driven initialization and pixel-wise densification strategy for Gaussian points to integrate new content. By combining all those techniques, GFlow transcends the boundaries of 4D recovery from causal videos; it naturally enables tracking of points and segmentation of moving objects across frames. Additionally, GFlow estimates the camera poses for each frame, enabling novel view synthesis by changing camera pose. This capability facilitates extensive scene-level or object-level editing, highlighting GFlow's versatility and effectiveness. Visit our project page at: https://littlepure2333.github.io/GFlow