UFMF: Radar-Guided Kinematic Priors for Real-Time 4D Occupancy Forecasting

Zeping Zhang; Zhuoya Zhao; Robert Laganière

cs.CVOct 29, 2026

UFMF: Radar-Guided Kinematic Priors for Real-Time 4D Occupancy Forecasting

Authors: Zeping Zhang, Zhuoya Zhao, Robert Laganière

Organizations: University of Ottawa

Abstract

Real-time 4D occupancy forecasting requires reliably extrapolating both ego motion and surrounding actor motion under strict latency budgets. Recent occupancy world models improve long-horizon fidelity through generative decoders and 3D volumetric modeling, yet their computational cost makes real-time deployment challenging. Light-weight BEV forecasters meet latency requirements but often blur or suppress fast-moving actors at longer horizons. We present UFMF (Ultra-Fast Multi-modal Forecaster), a real-time forecaster that keeps an attention-free diagonal state-space temporal core while injecting targeted kinematic and geometric structure where lightweight models typically fail. A radar-guided velocity-aware dynamic prior (VADP) decomposes future displacement into a linear extrapolation plus a learned horizon-dependent residual. The resulting per-cell flow fields transport a compact dynamic latent subspace during rollout. Two auxiliary training constraints complement this branch: world-frame dynamic supervision enforces inter-horizon trajectory coherence, and class-conditioned vertical-profile supervision regularizes BEV-to-3D lifting. UFMF operates at 122 FPS on a single RTX 4090 and reaches 0.45 fast-displacement mIoU on nuScenes, outperforming recent occupancy world models at a fraction of their computational cost.

Figures & tables

Figure 1: Real-data teaser at the last forecast horizon on a nuScenes validation window. (a) Our lightweight BEV baseline (i.e., the first row in Table 2 : shared backbone without VADP or auxiliary constraints) predicts only a partial future footprint for the boxed fast-moving car. (b) UFMF preserves a fuller, better-aligned footprint.

Figure 2: Overview of UFMF. Semantic history and rasterized radar cues are encoded by a lightweight BEV encoder and diagonal SSM history encoder, then propagated by a future rollout. VADP uses semantic deltas and radar cues to predict

\Delta_{k}=k\hat{\mathbf{v}}+r_{k}

and dynamic support

Q_{k}(c,p)

, which guide the rollout state before decoding. The prediction output panel shows the lifted voxel forecast, its BEV projection, and the training-only world-frame and height-profile supervision heads.

Method	mIoU(all) $\uparrow$	Static $\uparrow$	Slow $\uparrow$	Fast $\uparrow$	FPS $\uparrow$ / ms $\downarrow$
OccWorld [ 19 ]	0.478	0.571	0.472	0.336	23 / 44
Drive-OccWorld [ 17 ]	0.509	0.601	0.506	0.381	$\sim$ 2 / 647
COME [ 14 ]	0.527	0.618	0.526	0.402	$\sim$ 1 / 1157
UFMF (ours)	0.542	0.632	0.541	0.450	122 / 8

Table 1: Main comparison on a nuScenes [ 2 ] validation set. mIoU(all) averages over all semantic classes, including background. Static/Slow/Fast denote target-actor displacement buckets of

<2

2

–

10

m, and

\geq 10

m, respectively.

Figure 4: 3D qualitative comparison on one representative nuScenes [ 2 ] validation scene over time. Columns show Ground Truth(GT), OccWorld, Drive-OccWorld, COME, and UFMF, while rows correspond to $t{=}1$ s, $t{=}2$ s, and $t{=}3$ s from the same oblique camera viewpoint. UFMF preserves cleaner actor extent and more stable scene structure as the horizon increases, especially around elongated vehicles and sparse dynamic clusters.

Variant	mIoU(all) $\uparrow$	Static $\uparrow$	Slow $\uparrow$	Fast $\uparrow$	FPS $\uparrow$ / ms $\downarrow$
Lightweight BEV baseline	0.433	0.521	0.426	0.298	138 / 7
+ VADP	0.529	0.618	0.524	0.403	122 / 8
+ VADP + $\mathcal{L}_{\text{wf}}$	0.536	0.626	0.531	0.418	122 / 8
+ VADP + $\mathcal{L}_{\text{wf}}$ + $\mathcal{L}_{\text{traj}}$	0.539	0.631	0.538	0.435	122 / 8
+ $\mathcal{L}_{\text{height}}$ (Full UFMF)	0.542	0.632	0.541	0.450	122 / 8

Table 2: Ablation study on the proposed components. VADP mainly improves the fast-displacement bucket, world-frame constraints stabilize slow- and fast-displacement trajectories, and the height regularizer preserves the final overall mIoU without sacrificing runtime.

Figure 5: BEV comparison on one validation scene focused on a single tracked actor from the fast-displacement bucket over time. Columns show GT, OccWorld, Drive-OccWorld, COME, and UFMF, while rows correspond to $t{=}1$ s, $t{=}2$ s, and $t{=}3$ s. Each subfigure is centered on the same tracked object, and the black box marks the region of interest. UFMF better preserves compact support and temporal alignment, whereas the weaker rows show stronger fragmentation, shrinkage, or stray occupancy.

Radar condition	mIoU(all) $\uparrow$	Fast $\uparrow$	FPS $\uparrow$ / ms $\downarrow$
Full VADP (UFMF)	0.542	0.450	122/8
Noisy radar ( $\sigma{=}5$ m/s)	0.534	0.431	122/8
Radar-only VADP (no sem. delta)	0.535	0.420	122/8
Sem.-delta-only VADP (No radar)	0.529	0.407	128/8
Lightweight BEV baseline	0.433	0.298	138/7

Table 3: Radar modality isolation. The semantic-delta-only row removes radar from VADP; the radar-only row removes semantic deltas. Both retain the structural VADP mechanism, confirming that the architecture contributes independently of either input modality.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Method	Veh. IoU $\uparrow$	Horizon	FPS $\uparrow$	Dim.
FIERY [ 5 ]	36.7	2 s	28	2D
PowerBEV [ 9 ]	39.3	2 s	52	2D
UFMF (veh. class)	57.8	3 s	122	3D $\to$ 2D

Appendix

Table A1 : 2D vehicle IoU against efficient BEV forecasters on nuScenes. UFMF is evaluated at the harder 3 s horizon.

History source	mIoU(all) $\uparrow$	Fast $\uparrow$	FPS $\uparrow$
GT history (standard)	0.542	0.450	122
Predicted (BEVDet-Occ)	0.521	0.427	122

Appendix

Table B1 : Robustness to predicted (non-ground-truth) history inputs.

3D Occupancy Prediction

3D occupancy prediction estimates the geometry and semantics of occupied and free space around an agent from camera, LiDAR, or radar data.

See all 119 papers

FreeOcc: Decoupling Ego-Motion for Efficient 4D Occupancy Forecasting via Continuous Flow Matching
Nov 5, 2026Zeping Zhang, Zhuoya Zhao, Samy Metari +1World Models for Autonomous Driving 3D Occupancy Prediction
University of Ottawa · National Research Council Canada
InfraWorld: Motion-Aware World Modeling for Roadside 4D Occupancy Forecasting
Oct 10, 2026Songkai Wang, Xiaokai Bai, Lei Yang +1Spatiotemporal Forecasting 3D Occupancy Prediction
College of Information Science and Electronic Engineering, Zhejiang University. · School of Mechanical and Aerospace Engineering, Nanyang Technological University.
InterOCF: Spatio-Temporal 2D-3D Interaction for Camera-Only 4D Occupancy Forecasting
Jul 27, 2026Qi Zhang, Xinquan Yu, Kaiyi Zhang +13D Occupancy Prediction Spatiotemporal Forecasting
College of Computer Science and Software Engineering, Shenzhen University, Shenzhen, China.

Nov 5, 2026J/K move · Enter open · S save

FreeOcc: Decoupling Ego-Motion for Efficient 4D Occupancy Forecasting via Continuous Flow Matching

Zeping Zhang, Zhuoya Zhao, Samy Metari, Robert Laganière

University of Ottawa · National Research Council Canada

World Models for Autonomous Driving 3D Occupancy Prediction

Open paper AskPDF arXiv

Current 4D occupancy world models use heavy cross-attention or deep recurrence to disentangle ego-motion from scene dynamics. This requires significant computational capacity. However, this effort is largely redundant since the ego-vehicle's intended future motion is provided by the downstream motion planner. We present FreeOcc, a lightweight continuous-time world model that introduces explicit

SE(2)

geometric decoupling into a conditional flow-matching framework. By mathematically factoring out self-motion, we create a purely dynamic feature space. This space supports attention-free temporal fusion and allows physics-constrained ODEs to operate strictly on residual agent kinematics. To maximize deployment efficiency, FreeOcc introduces an uncertainty-aware blending mechanism. This routes well-observed voxels to a single-pass decoder, reserving the ODE solver exclusively for ambiguous, high-uncertainty regions. Requiring only 28.3M parameters-roughly 40% of comparable models-our optimal ensemble mode (FreeOcc-E) runs at 27 FPS. On nuScenes, FreeOcc achieves state-of-the-art forecasting performance with a 0.47 m Near Field Chamfer Distance (NFCD) and 32.1% mIoU. Furthermore, the model transfers zero-shot to SemanticKITTI. It also proves highly robust, retaining 92.8% of its ideal mIoU under significant sensor packet loss and timestamp jitter.

Abstract

3D Occupancy Prediction

Explore similar work

FreeOcc: Decoupling Ego-Motion for Efficient 4D Occupancy Forecasting via Continuous Flow Matching

InfraWorld: Motion-Aware World Modeling for Roadside 4D Occupancy Forecasting

InterOCF: Spatio-Temporal 2D-3D Interaction for Camera-Only 4D Occupancy Forecasting