DiST-4D: Disentangled Spatiotemporal Diffusion with Metric Depth for 4D Driving Scene Generation

• URL: http://arxiv.org/abs/2503.15208v1
• Date: Wed, 19 Mar 2025 13:49:48 GMT
• Title: DiST-4D: Disentangled Spatiotemporal Diffusion with Metric Depth for 4D Driving Scene Generation
• Authors: Jiazhe Guo, Yikang Ding, Xiwu Chen, Shuo Chen, Bohan Li, Yingshuang Zou, Xiaoyang Lyu, Feiyang Tan, Xiaojuan Qi, Zhiheng Li, Hao Zhao,
• Abstract summary: Current generative models struggle to synthesize 4D driving scenes that simultaneously support temporal extrapolation and spatial novel view synthesis (NVS)<n>We propose DiST-4D, which disentangles the problem into two diffusion processes: DiST-T, which predicts future metric depth and multi-view RGB sequences directly from past observations, and DiST-S, which enables spatial NVS by training only on existing viewpoints while enforcing cycle consistency.<n>Experiments demonstrate that DiST-4D achieves state-of-the-art performance in both temporal prediction and NVS tasks, while also delivering competitive performance in planning-related evaluations.
• Score: 50.01520547454224
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Current generative models struggle to synthesize dynamic 4D driving scenes that simultaneously support temporal extrapolation and spatial novel view synthesis (NVS) without per-scene optimization. A key challenge lies in finding an efficient and generalizable geometric representation that seamlessly connects temporal and spatial synthesis. To address this, we propose DiST-4D, the first disentangled spatiotemporal diffusion framework for 4D driving scene generation, which leverages metric depth as the core geometric representation. DiST-4D decomposes the problem into two diffusion processes: DiST-T, which predicts future metric depth and multi-view RGB sequences directly from past observations, and DiST-S, which enables spatial NVS by training only on existing viewpoints while enforcing cycle consistency. This cycle consistency mechanism introduces a forward-backward rendering constraint, reducing the generalization gap between observed and unseen viewpoints. Metric depth is essential for both accurate reliable forecasting and accurate spatial NVS, as it provides a view-consistent geometric representation that generalizes well to unseen perspectives. Experiments demonstrate that DiST-4D achieves state-of-the-art performance in both temporal prediction and NVS tasks, while also delivering competitive performance in planning-related evaluations.

Related papers

• Orthogonal Spatial-temporal Distributional Transfer for 4D Generation [58.30004328671699]
  We propose a framework that transfers rich spatial priors from existing 3D diffusion models and temporal priors from video diffusion models to enhance 4D synthesis.<n>We develop a spatial-temporal-disentangled 4D (STD-4D) Diffusion model, which synthesizes 4D-aware videos through disentangled spatial and temporal latents.<n> Experiments demonstrate that our method significantly outperforms existing approaches, achieving superior spatial-temporal consistency and higher-quality 4D synthesis.
  arXiv  Detail & Related papers  (2026-03-05T11:52:21Z)
• From Frames to Sequences: Temporally Consistent Human-Centric Dense Prediction [22.291273919939957]
  We develop a scalable synthetic data pipeline that generates human frames and motion-aligned sequences with pixel-accurate depth, normals, and masks.<n>We train a unified ViT-based dense predictor that injects an explicit geometric human prior via CSE embeddings.<n>Our two-stage training strategy, combining static pretraining with dynamic sequence supervision, enables the model first to acquire robust spatial representations and then refine temporal consistency across motion-aligned sequences.
  arXiv  Detail & Related papers  (2026-02-02T05:28:58Z)
• Tracking-Guided 4D Generation: Foundation-Tracker Motion Priors for 3D Model Animation [21.075786141331974]
  We present emphTrack4DGen, a framework for generating dynamic 4D objects from sparse inputs.<n>In Stage One, we enforce dense, feature-level point correspondences inside the diffusion generator.<n>In Stage Two, we reconstruct a dynamic 4D-GS using a hybrid motion encoding.
  arXiv  Detail & Related papers  (2025-12-05T21:13:04Z)
• U4D: Uncertainty-Aware 4D World Modeling from LiDAR Sequences [54.77163447282599]
  Existing generative frameworks treat all spatial regions uniformly, overlooking the varying uncertainty across real-world scenes.<n>We present U4D, an uncertainty-aware framework for 4D LiDAR world modeling.<n>Our approach first estimates spatial uncertainty maps from a pretrained segmentation model to localize semantically challenging regions.<n>It then performs generation in a "hard-to-easy" manner through two sequential stages: (1) uncertainty-region modeling, which reconstructs high-entropy regions with fine geometric fidelity, and (2) uncertainty-conditioned completion, which synthesizes the remaining areas under learned structural priors.
  arXiv  Detail & Related papers  (2025-12-02T17:59:57Z)
• 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)
• 4D Driving Scene Generation With Stereo Forcing [62.47705572424127]
  Current generative models struggle to synthesize dynamic 4D driving scenes that simultaneously support temporal extrapolation and spatial novel view synthesis (NVS) without per-scene optimization.<n>We present PhiGenesis, a unified framework for 4D scene generation that extends video generation techniques with geometric and temporal consistency.
  arXiv  Detail & Related papers  (2025-09-24T15:37:17Z)
• Dream4D: Lifting Camera-Controlled I2V towards Spatiotemporally Consistent 4D Generation [3.1852855132066673]
  Current approaches often struggle to maintain view consistency while handling complex scene dynamics.<n>This framework is the first to leverage both rich temporal priors video diffusion models and geometric awareness of the reconstruction models.<n>It significantly facilitates 4D generation and shows higher quality (e.g., mPSNR, mSSIM) over existing methods.
  arXiv  Detail & Related papers  (2025-08-11T08:55:47Z)
• STP4D: Spatio-Temporal-Prompt Consistent Modeling for Text-to-4D Gaussian Splatting [34.07501669897291]
  Text-to-4D generation is rapidly developing widely applied in various scenarios. Existing methods often fail to incorporate adequate-prompt modeling within a unified framework. We propose a novel approach to integrate comprehensive texts-to-4D generation.
  arXiv  Detail & Related papers  (2025-04-25T12:53:15Z)
• Free4D: Tuning-free 4D Scene Generation with Spatial-Temporal Consistency [49.875459658889355]
  Free4D is a tuning-free framework for 4D scene generation from a single image. Our key insight is to distill pre-trained foundation models for consistent 4D scene representation. The resulting 4D representation enables real-time, controllable rendering.
  arXiv  Detail & Related papers  (2025-03-26T17:59:44Z)
• Learning 4D Panoptic Scene Graph Generation from Rich 2D Visual Scene [122.42861221739123]
  This paper investigates a novel framework for 4D-PSG generation that leverages rich 2D visual scene annotations to enhance 4D scene learning.<n>We propose a 2D-to-4D visual scene transfer learning framework, where a spatial-temporal scene strategy effectively transfers dimension-invariant features from abundant 2D SG annotations to 4D scenes.
  arXiv  Detail & Related papers  (2025-03-19T09:16:08Z)
• 4D Gaussian Splatting: Modeling Dynamic Scenes with Native 4D Primitives [116.2042238179433]
  In this paper, we frame dynamic scenes as unconstrained 4D volume learning problems.<n>We represent a target dynamic scene using a collection of 4D Gaussian primitives with explicit geometry and appearance features.<n>This approach can capture relevant information in space and time by fitting the underlying photorealistic-temporal volume.<n> Notably, our 4DGS model is the first solution that supports real-time rendering of high-resolution, novel views for complex dynamic scenes.
  arXiv  Detail & Related papers  (2024-12-30T05:30:26Z)
• 4DRecons: 4D Neural Implicit Deformable Objects Reconstruction from a single RGB-D Camera with Geometrical and Topological Regularizations [35.161541396566705]
  4DRecons encodes the output as a 4D neural implicit surface. We show that 4DRecons can handle large deformations and complex inter-part interactions.
  arXiv  Detail & Related papers  (2024-06-14T16:38:00Z)
• PLA4D: Pixel-Level Alignments for Text-to-4D Gaussian Splatting [9.517058280333806]
  Previous text-to-4D methods have leveraged multiple Score Distillation Sampling (SDS) techniques. We introduce textbfPixel-textbfLevel textbfAlignment for text-driven textbf4D Gaussian splatting (PLA4D) PLA4D provides an anchor reference, i.e., text-generated video, to align the rendering process conditioned by different DMs in pixel space.
  arXiv  Detail & Related papers  (2024-05-30T11:23:01Z)
• Diffusion4D: Fast Spatial-temporal Consistent 4D Generation via Video Diffusion Models [116.31344506738816]
  We present a novel framework, textbfDiffusion4D, for efficient and scalable 4D content generation. We develop a 4D-aware video diffusion model capable of synthesizing orbital views of dynamic 3D assets. Our method surpasses prior state-of-the-art techniques in terms of generation efficiency and 4D geometry consistency.
  arXiv  Detail & Related papers  (2024-05-26T17:47:34Z)
• A Spatiotemporal Approach to Tri-Perspective Representation for 3D Semantic Occupancy Prediction [6.527178779672975]
  Vision-based 3D semantic occupancy prediction is increasingly overlooked in favor of LiDAR-based approaches.<n>This study introduces S2TPVFormer, a transformer architecture designed to predict temporally coherent 3D semantic occupancy.
  arXiv  Detail & Related papers  (2024-01-24T20:06:59Z)
• Motion2VecSets: 4D Latent Vector Set Diffusion for Non-rigid Shape Reconstruction and Tracking [52.393359791978035]
  Motion2VecSets is a 4D diffusion model for dynamic surface reconstruction from point cloud sequences. We parameterize 4D dynamics with latent sets instead of using global latent codes. For more temporally-coherent object tracking, we synchronously denoise deformation latent sets and exchange information across multiple frames.
  arXiv  Detail & Related papers  (2024-01-12T15:05:08Z)
• LoRD: Local 4D Implicit Representation for High-Fidelity Dynamic Human Modeling [69.56581851211841]
  We propose a novel Local 4D implicit Representation for Dynamic clothed human, named LoRD. Our key insight is to encourage the network to learn the latent codes of local part-level representation. LoRD has strong capability for representing 4D human, and outperforms state-of-the-art methods on practical applications.
  arXiv  Detail & Related papers  (2022-08-18T03:49:44Z)
• Learning Parallel Dense Correspondence from Spatio-Temporal Descriptors for Efficient and Robust 4D Reconstruction [43.60322886598972]
  This paper focuses on the task of 4D shape reconstruction from a sequence of point clouds. We present a novel pipeline to learn a temporal evolution of the 3D human shape through capturing continuous transformation functions among cross-frame occupancy fields.
  arXiv  Detail & Related papers  (2021-03-30T13:36:03Z)

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.