PointForward: Feedforward Driving Reconstruction through Point-Aligned Representations

• URL: http://arxiv.org/abs/2605.11594v1
• Date: Tue, 12 May 2026 06:20:59 GMT
• Title: PointForward: Feedforward Driving Reconstruction through Point-Aligned Representations
• Abstract summary: 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.
• Score: 39.816194543559526
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: High-fidelity reconstruction of driving scenes is crucial for autonomous driving. While recent feedforward 3D Gaussian Splatting (3DGS) methods enable fast reconstruction, their per-pixel Gaussian prediction paradigm often suffers from multi-view inconsistency and layering artifacts. Moreover, existing methods often model dynamic instances via dense flow prediction, which lacks explicit cross-view correspondence and instance-level consistency. In this paper, we propose PointForward, a feedforward driving reconstruction framework through point-aligned representations. Unlike pixel-aligned methods, we initialize sparse 3D queries in world space and aggregate multi-view image information via spatial-temporal fusion onto these queries, enforcing explicit cross-view consistency in a single feedforward pass. To handle scene dynamics, we introduce scene graphs that explicitly organize moving instances during reconstruction. By leveraging 3D bounding boxes, our method enables instance-level motion propagation and temporally consistent dynamic representations. Extensive experiments demonstrate that PointForward achieves state-of-the-art performance on large-scale driving benchmarks. The code will be available upon the publication of the paper.

Related papers

• USR-Drive: Unified Driving Scene Representation via Joint Denoising of 3D Gaussians and Boxes [57.4583999658488]
  Spatial representation learning for autonomous driving aims to map raw visual signals into structured 3D scene representations.<n>Existing methods treat dynamic reconstruction and instance-level perception as separate tasks.<n>We propose USR-Drive, a unified conditional generative framework that recovers dense dynamic geometry and instance-level object layouts within a shared scene representation.
  arXiv  Detail & Related papers  (2026-08-19T15:29:06Z)
• InfiniVerse: Occupancy Guided Unbounded Scene Generation for Autonomous Driving [70.7737035270887]
  We introduce InfiniVerse, a unified pipeline for synthesis of dynamic urban scenes from a single frame.<n>InfiniVerse achieves state-of-the-art performance with a FID of 6.4 and FVD of 67.97, significantly outperforming existing benchmarks in both duration and stability.
  arXiv  Detail & Related papers  (2026-06-30T04:08:22Z)
• DrivingVoxels: Compositional Sparse Voxel Rasterization for Dynamic Driving Scene Reconstruction [14.8989510421373]
  We present DrivingVoxels, a compositional sparse voxel rendering framework for dynamic driving scenes.<n>Our method jointlyizes sparse voxels from multiple independent octrees within a single rendering pass.<n>We evaluate our framework on the PandaSet benchmark, demonstrating that DrivingVoxels performs on par on perceptual metrics and better on structural metrics for NVS and reconstruction.
  arXiv  Detail & Related papers  (2026-06-22T08:42:16Z)
• FAST3DIS: Feed-forward Anchored Scene Transformer for 3D Instance Segmentation [15.271467111162714]
  FAST3DIS is an end-to-end approach that effectively bypasses post-hoc clustering.<n>We introduce a 3D-anchored, query-based Transformer architecture built upon a foundational depth backbone.<n> Experiments on complex indoor 3D datasets demonstrate that our approach achieves competitive segmentation accuracy.
  arXiv  Detail & Related papers  (2026-03-27T00:45:31Z)
• SegVGGT: Joint 3D Reconstruction and Instance Segmentation from Multi-View Images [11.617237358347777]
  SegVGGT is a unified end-to-end framework that simultaneously performs feed-forward 3D reconstruction and instance segmentation.<n>Our method deeply integrates instance identification into the visual geometry grounded transformer.<n>Experiments demonstrate that SegVGGT achieves the state-of-the-art performance on ScanNetv2 and ScanNet200.
  arXiv  Detail & Related papers  (2026-03-20T13:10:46Z)
• StreetForward: Perceiving Dynamic Street with Feedforward Causal Attention [11.60018084377282]
  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.
  arXiv  Detail & Related papers  (2026-03-20T01:31:05Z)
• IDSplat: Instance-Decomposed 3D Gaussian Splatting for Driving Scenes [25.939318593012484]
  Reconstructing dynamic driving scenes is essential for developing autonomous systems through sensor-realistic simulation.<n>We present IDSplat, a self-supervised 3D Gaussian Splatting framework that reconstructs dynamic scenes with explicit instance decomposition and learnable motion trajectories.<n>Our method achieves competitive reconstruction quality while maintaining instance-level decomposition and generalizes across diverse sequences and view densities without retraining.
  arXiv  Detail & Related papers  (2025-11-24T15:48:08Z)
• POMATO: Marrying Pointmap Matching with Temporal Motion for Dynamic 3D Reconstruction [53.19968902152528]
  We present POMATO, a unified framework for dynamic 3D reconstruction by marrying pointmap matching with temporal motion.<n>Specifically, our method learns an explicit matching relationship by mapping RGB pixels from both dynamic and static regions across different views to 3D pointmaps.<n>We show the effectiveness of the proposed pointmap matching and temporal fusion paradigm by demonstrating the remarkable performance across multiple downstream tasks.
  arXiv  Detail & Related papers  (2025-04-08T05:33:13Z)
• Driv3R: Learning Dense 4D Reconstruction for Autonomous Driving [116.10577967146762]
  We propose Driv3R, a framework that directly regresses per-frame point maps from multi-view image sequences.<n>We employ a 4D flow predictor to identify moving objects within the scene to direct our network focus more on reconstructing these dynamic regions.<n>Driv3R outperforms previous frameworks in 4D dynamic scene reconstruction, achieving 15x faster inference speed.
  arXiv  Detail & Related papers  (2024-12-09T18:58:03Z)
• Modeling Continuous Motion for 3D Point Cloud Object Tracking [54.48716096286417]
  This paper presents a novel approach that views each tracklet as a continuous stream. At each timestamp, only the current frame is fed into the network to interact with multi-frame historical features stored in a memory bank. To enhance the utilization of multi-frame features for robust tracking, a contrastive sequence enhancement strategy is proposed.
  arXiv  Detail & Related papers  (2023-03-14T02:58:27Z)
• FIRe: Fast Inverse Rendering using Directional and Signed Distance Functions [97.5540646069663]
  We introduce a novel neural scene representation that we call the directional distance function (DDF) Our DDF is defined on the unit sphere and predicts the distance to the surface along any given direction. Based on our DDF, we present a novel fast algorithm (FIRe) to reconstruct 3D shapes given a posed depth map.
  arXiv  Detail & Related papers  (2022-03-30T13:24:04Z)

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.