DiST-4D: Disentangled Spatiotemporal Diffusion with Metric Depth for 4D Driving Scene Generation
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:HS6GJMBNrecord.jsonopen to challenge →
read the original 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.
This paper has not been read by Pith yet.
Forward citations
Cited by 6 Pith papers
-
From Articulated Kinematics to Routed Visual Control for Action-Conditioned Surgical Video Generation
A kinematic-to-visual lifting paradigm combined with hierarchically routed control generates action-conditioned surgical videos with better faithfulness, fidelity, and efficiency.
-
Point as Skeleton: Accumulated Point Cloud Enhanced Autoregressive Generation for Closed-Loop Autonomous Driving Simulation
Point-cloud skeleton conditions and a Reset-and-Roll inference scheme enable stable frame-wise autoregressive driving video generation for closed-loop autonomous driving simulation.
-
Relit-LiVE: Relight Video by Jointly Learning Environment Video
Relit-LiVE jointly predicts relit videos and viewpoint-aligned environment maps inside a single diffusion process to achieve physically consistent video relighting without camera pose input.
-
DriveLaW:Unifying Planning and Video Generation in a Latent Driving World
DriveLaW unifies video world modeling and trajectory planning by injecting video-generator latents into a diffusion planner, achieving SOTA video prediction and a new record on the NAVSIM planning benchmark.
-
GaussianDWM: 3D Gaussian Driving World Model for Unified Scene Understanding and Multi-Modal Generation
GaussianDWM uses 3D Gaussians with embedded linguistic features, language-guided sampling, and dual-condition generation for unified scene understanding and multi-modal output in driving world models.
-
FrozenDrive: Zero-Shot Text-Guided Driving Scene Generation and Data Augmentation with Parameter-Free Frozen Diffusion Model
FrozenDrive enables zero-shot text-guided generation of consistent multi-view driving scenes via a parameter-free frozen diffusion backbone with spatio-temporal attention, improving autonomous driving models on advers...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.