DrivingScene: A Multi-Task Online Feed-Forward 3D Gaussian Splatting Method for Dynamic Driving Scenes
Real-time, high-fidelity reconstruction of dynamic driving scenes is challenged by complex dynamics and sparse views, with prior methods struggling to balance quality and efficiency. We propose DrivingScene, an online, feed-forward framework that reconstructs 4D dynamic scenes from only two consecutive surround-view images. Our key innovation is a lightweight residual flow network that predicts the non-rigid motion of dynamic objects per camera on top of a learned static scene prior, explicitly modeling dynamics via scene flow. We also introduce a coarse-to-fine training paradigm that circumvents the instabilities common to end-to-end approaches. Experiments on nuScenes dataset show our image-only method simultaneously generates high-quality depth, scene flow, and 3D Gaussian point clouds online, significantly outperforming state-of-the-art methods in both dynamic reconstruction and novel view synthesis.
Code (0)
등록된 구현이 없습니다.
Tasks
Dynamic ReconstructionNovel View SynthesisPoint CloudsSimilar Papers 제목 키워드 기반
Feedforward Control in the Presence of Input Nonlinearities: A Learning-based Approach
Advanced feedforward control methods enable mechatronic systems to perform varying motion tasks with extreme accuracy and throughput. The aim of this paper is to develop a data-driven feedforward controller that addresse…
Learning nonlinear feedforward: a Gaussian Process Approach Applied to a Printer with Friction
Feedforward control is essential to achieving good tracking performance in positioning systems. The aim of this paper is to develop an identification strategy for inverse models of systems with nonlinear dynamics of unkn…
FrictionOn the Role of Transformer Feed-Forward Layers in Nonlinear In-Context Learning
Transformer-based models demonstrate a remarkable ability for in-context learning (ICL), where they can adapt to unseen tasks from a few prompt examples without parameter updates. Notably, recent research has provided in…
In-Context LearningPhysics-Guided Neural Networks for Feedforward Control: An Orthogonal Projection-Based Approach
Unknown nonlinear dynamics can limit the performance of model-based feedforward control. The aim of this paper is to develop a feedforward control framework for systems with unknown, typically nonlinear, dynamics. To add…
FrictionFreeStreamGS: Online Feed-forward 3D Gaussian Splatting from Unposed Streaming Inputs
Feed-forward 3D Gaussian Splatting (3DGS) allows efficient and high-fidelity novel view synthesis (NVS) from an offline recorded image sequence. However, achieving online NVS from streaming and unposed image inputs remai…
Novel View Synthesis