Depth Anything V4: Dynamic 4D Scene Reconstruction via Riemannian Flow Matching on 4D Gaussian Splatting

• URL: http://arxiv.org/abs/2608.18388v2
• Date: Thu, 20 Aug 2026 09:04:46 GMT
• Title: Depth Anything V4: Dynamic 4D Scene Reconstruction via Riemannian Flow Matching on 4D Gaussian Splatting
• Abstract summary: We present Depth Anything V4 (DAV4), a framework for dynamic 4D scene reconstruction from monocular video.<n>Key contribution is the application of Riemannian Flow Matching (RFM) to 4D Gaussian Splatting parameters.<n>DAV4 outperforms prior Depth Anything models and per-scene 4D-GS on dynamic reconstruction and novel-view synthesis.
• Score: 9.090610968599192
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: We present Depth Anything V4 (DAV4), a framework for dynamic 4D scene reconstruction from monocular video. Our key contribution is the application of Riemannian Flow Matching (RFM) to 4D Gaussian Splatting parameters, defining probability paths directly on non-Euclidean manifolds (scale, rotation, opacity), ensuring all intermediate states are valid. Through controlled experiments, we isolate RFM's contribution from test-time optimization (TTO) and pre-training. A deterministic MLP baseline with the same data, architecture, and TTO achieves F-score 0.762; RFM achieves 0.806 - the +0.044 gain is RFM's isolated contribution. We provide corrected computational cost analysis: pre-training is 360 GPU-hours, amortizing for large-scale deployment (over 10,000 scenes). Uncertainty is quantified via Negative Gaussian Log-Likelihood and Expected Calibration Error. DAV4 outperforms prior Depth Anything models and per-scene 4D-GS on dynamic reconstruction and novel-view synthesis, while using no human-annotated depth labels as training losses.

Related papers

• DepthART: Scaling Foundation Monocular Depth to Tiny Models [51.10174763031444]
  DepthART is a compact MDE model for on-device deployment across diverse scenes.<n>We show how DepthART consistently surpasses previous tiny baselines in both zero-shot generalization and metric accuracy.<n>We also provide a scalable model family, with DepthART-S reaching 347/245 FPS (strict FP32) on an GTX A6000 at $2242/4482$, 102 FPS (TF32) on a Orin NX 8GB, and over 15 FPS (FP32) on a Jetson Nano 4GB.
  arXiv  Detail & Related papers  (2026-07-19T06:54:52Z)
• Geometry-Consistent 4D Gaussian Splatting for Sparse-Input Dynamic View Synthesis [17.560425604804305]
  GC-4DGS is a novel framework that infuses geometric consistency into 4D Gaussian Splatting (4DGS)<n>This paper presents GC-4DGS, a novel framework that infuses geometric consistency into 4D Gaussian Splatting (4DGS)
  arXiv  Detail & Related papers  (2025-11-28T10:11:48Z)
• Diff4Splat: Controllable 4D Scene Generation with Latent Dynamic Reconstruction Models [79.06910348413861]
  We introduce Diff4Splat, a feed-forward method that synthesizes controllable and explicit 4D scenes from a single image.<n>Given a single input image, a camera trajectory, and an optional text prompt, Diff4Splat directly predicts a deformable 3D Gaussian field that encodes appearance, geometry, and motion.
  arXiv  Detail & Related papers  (2025-11-01T11:16:25Z)
• Optimized Minimal 4D Gaussian Splatting [50.29519144753797]
  4D Gaussian Splatting has emerged as a new paradigm for dynamic scene representation, enabling real-time rendering of scenes with complex motions.<n>It faces a major challenge of storage overhead, as millions of Gaussians are required for high-fidelity reconstruction.<n>We present OMG4, a framework that constructs a compact set of salient Gaussians capable of faithfully representing 4D Gaussian models.
  arXiv  Detail & Related papers  (2025-10-04T16:11:13Z)
• 4DRGS: 4D Radiative Gaussian Splatting for Efficient 3D Vessel Reconstruction from Sparse-View Dynamic DSA Images [49.170407434313475]
  Existing methods often produce suboptimal results or require excessive computation time.<n>We propose 4D radiative Gaussian splatting (4DRGS) to achieve high-quality reconstruction efficiently.<n>4DRGS achieves impressive results in 5 minutes training, which is 32x faster than the state-of-the-art method.
  arXiv  Detail & Related papers  (2024-12-17T13:51:56Z)
• GPS-Gaussian+: Generalizable Pixel-wise 3D Gaussian Splatting for Real-Time Human-Scene Rendering from Sparse Views [67.34073368933814]
  We propose a generalizable Gaussian Splatting approach for high-resolution image rendering under a sparse-view camera setting. We train our Gaussian parameter regression module on human-only data or human-scene data, jointly with a depth estimation module to lift 2D parameter maps to 3D space. Experiments on several datasets demonstrate that our method outperforms state-of-the-art methods while achieving an exceeding rendering speed.
  arXiv  Detail & Related papers  (2024-11-18T08:18:44Z)
• Self-Calibrating 4D Novel View Synthesis from Monocular Videos Using Gaussian Splatting [14.759265492381509]
  We propose a novel approach that learns a high-fidelity 4D GS scene representation with self-calibration of camera parameters.<n>It includes the extraction of 2D point features that robustly represent 3D structure.<n>Results show significant improvements over state-of-the-art methods for 4D novel view synthesis.
  arXiv  Detail & Related papers  (2024-06-03T06:52:35Z)
• Q-SLAM: Quadric Representations for Monocular SLAM [85.82697759049388]
  We reimagine volumetric representations through the lens of quadrics. We use quadric assumption to rectify noisy depth estimations from RGB inputs. We introduce a novel quadric-decomposed transformer to aggregate information across quadrics.
  arXiv  Detail & Related papers  (2024-03-12T23:27:30Z)
• Fully Differentiable and Interpretable Model for VIO with 4 Trainable Parameters [16.347927939872488]
  Monocular visual-inertial odometry is a critical problem in robotics and autonomous driving. In this paper, we propose a fully differentiable, interpretable, and lightweight monocular VIO model that contains only 4 trainable parameters. Experimental results on synthetic and real-world datasets demonstrate that our simple approach is competitive with state-of-the-art methods.
  arXiv  Detail & Related papers  (2021-09-25T06:54:09Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.