StreetForward: Perceiving Dynamic Street with Feedforward Causal Attention

• URL: http://arxiv.org/abs/2603.19552v1
• Date: Fri, 20 Mar 2026 01:31:05 GMT
• Title: StreetForward: Perceiving Dynamic Street with Feedforward Causal Attention
• Abstract summary: We present StreetForward, a pose-free and tracker-free feedforward framework for dynamic street reconstruction.<n>We propose a simple yet effective temporal mask attention module that captures dynamic motion information from image sequences.<n>We train and evaluate our model on the Open dataset, demonstrating superior performance on novel view and depth compared to existing methods.
• Score: 11.60018084377282
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Feedforward reconstruction is crucial for autonomous driving applications, where rapid scene reconstruction enables efficient utilization of large-scale driving datasets in closed-loop simulation and other downstream tasks, eliminating the need for time-consuming per-scene optimization. We present StreetForward, a pose-free and tracker-free feedforward framework for dynamic street reconstruction. Building upon the alternating attention mechanism from Visual Geometry Grounded Transformer (VGGT), we propose a simple yet effective temporal mask attention module that captures dynamic motion information from image sequences and produces motion-aware latent representations. Static content and dynamic instances are represented uniformly with 3D Gaussian Splatting, and are optimized jointly by cross-frame rendering with spatio-temporal consistency, allowing the model to infer per-pixel velocities and produce high-fidelity novel views at new poses and times. We train and evaluate our model on the Waymo Open Dataset, demonstrating superior performance on novel view synthesis and depth estimation compared to existing methods. Furthermore, zero-shot inference on CARLA and other datasets validates the generalization capability of our approach. More visualizations are available on our project page: https://streetforward.github.io.

Related papers

• PointForward: Feedforward Driving Reconstruction through Point-Aligned Representations [39.816194543559526]
  We propose PointForward, a feedforward driving reconstruction framework through point-aligned representations.<n>Unlike pixel-aligned methods, we aggregate multi-view image information via spatial-temporal fusion.<n>Our method enables instance-level motion propagation and temporally consistent dynamic representations.
  arXiv  Detail & Related papers  (2026-05-12T06:20:59Z)
• DGGT: Feedforward 4D Reconstruction of Dynamic Driving Scenes using Unposed Images [36.562825380568384]
  We introduce textbfDriving Gaussian Grounded Transformer (DGGT), a unified framework for pose-free dynamic scene reconstruction.<n>Our approach jointly predicts per-frame 3D Gaussian maps and camera parameters, disentangles dynamics with a lightweight dynamic head.<n>A diffusion-based rendering refinement further reduces motion/interpolation artifacts and improves novel-view quality under sparse inputs.
  arXiv  Detail & Related papers  (2025-12-02T18:29:18Z)
• Tracktention: Leveraging Point Tracking to Attend Videos Faster and Better [61.381599921020175]
  Temporal consistency is critical in video prediction to ensure that outputs are coherent and free of artifacts.<n>Traditional methods, such as temporal attention and 3D convolution, may struggle with significant object motion.<n>We propose the Tracktention Layer, a novel architectural component that explicitly integrates motion information using point tracks.
  arXiv  Detail & Related papers  (2025-03-25T17:58:48Z)
• An Efficient Occupancy World Model via Decoupled Dynamic Flow and Image-assisted Training [50.71892161377806]
  DFIT-OccWorld is an efficient 3D occupancy world model that leverages decoupled dynamic flow and image-assisted training strategy.<n>Our model forecasts future dynamic voxels by warping existing observations using voxel flow, whereas static voxels are easily obtained through pose transformation.
  arXiv  Detail & Related papers  (2024-12-18T12:10:33Z)
• MonST3R: A Simple Approach for Estimating Geometry in the Presence of Motion [118.74385965694694]
  We present Motion DUSt3R (MonST3R), a novel geometry-first approach that directly estimates per-timestep geometry from dynamic scenes.<n>By simply estimating a pointmap for each timestep, we can effectively adapt DUST3R's representation, previously only used for static scenes, to dynamic scenes.<n>We show that by posing the problem as a fine-tuning task, identifying several suitable datasets, and strategically training the model on this limited data, we can surprisingly enable the model to handle dynamics.
  arXiv  Detail & Related papers  (2024-10-04T18:00:07Z)
• D-NPC: Dynamic Neural Point Clouds for Non-Rigid View Synthesis from Monocular Video [53.83936023443193]
  This paper contributes to the field by introducing a new synthesis method for dynamic novel view from monocular video, such as smartphone captures.<n>Our approach represents the as a $textitdynamic neural point cloud$, an implicit time-conditioned point cloud that encodes local geometry and appearance in separate hash-encoded neural feature grids.
  arXiv  Detail & Related papers  (2024-06-14T14:35:44Z)
• Gear-NeRF: Free-Viewpoint Rendering and Tracking with Motion-aware Spatio-Temporal Sampling [70.34875558830241]
  We present a way for learning a-temporal (4D) embedding, based on semantic semantic gears to allow for stratified modeling of dynamic regions of rendering the scene. At the same time, almost for free, our tracking approach enables free-viewpoint of interest - a functionality not yet achieved by existing NeRF-based methods.
  arXiv  Detail & Related papers  (2024-06-06T03:37:39Z)
• Degrees of Freedom Matter: Inferring Dynamics from Point Trajectories [28.701879490459675]
  We aim to learn an implicit motion field parameterized by a neural network to predict the movement of novel points within same domain. We exploit intrinsic regularization provided by SIREN, and modify the input layer to produce atemporally smooth motion field. Our experiments assess the model's performance in predicting unseen point trajectories and its application in temporal mesh alignment with deformation.
  arXiv  Detail & Related papers  (2024-06-05T21:02:10Z)
• EmerNeRF: Emergent Spatial-Temporal Scene Decomposition via Self-Supervision [85.17951804790515]
  EmerNeRF is a simple yet powerful approach for learning spatial-temporal representations of dynamic driving scenes. It simultaneously captures scene geometry, appearance, motion, and semantics via self-bootstrapping. Our method achieves state-of-the-art performance in sensor simulation.
  arXiv  Detail & Related papers  (2023-11-03T17:59:55Z)
• Dyna-DepthFormer: Multi-frame Transformer for Self-Supervised Depth Estimation in Dynamic Scenes [19.810725397641406]
  We propose a novel Dyna-Depthformer framework, which predicts scene depth and 3D motion field jointly. Our contributions are two-fold. First, we leverage multi-view correlation through a series of self- and cross-attention layers in order to obtain enhanced depth feature representation. Second, we propose a warping-based Motion Network to estimate the motion field of dynamic objects without using semantic prior.
  arXiv  Detail & Related papers  (2023-01-14T09:43:23Z)

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.