Reconstruction Matters: Learning Geometry-Aligned BEV Representation through 3D Gaussian Splatting

TL;DR

This paper introduces Splat2BEV, a framework that enhances Bird's-Eye-View perception for autonomous driving by explicitly reconstructing 3D scenes with Gaussian Splatting, leading to more accurate and interpretable BEV representations.

Contribution

It proposes a novel approach that explicitly reconstructs 3D scenes using Gaussian Splatting to improve BEV perception accuracy and interpretability.

Findings

Achieves state-of-the-art performance on nuScenes and argoverse datasets.

Demonstrates the effectiveness of explicit 3D reconstruction for BEV tasks.

Validates the importance of geometry-aligned features in perception accuracy.

Abstract

Bird's-Eye-View (BEV) perception serves as a cornerstone for autonomous driving, offering a unified spatial representation that fuses surrounding-view images to enable reasoning for various downstream tasks, such as semantic segmentation, 3D object detection, and motion prediction. However, most existing BEV perception frameworks adopt an end-to-end training paradigm, where image features are directly transformed into the BEV space and optimized solely through downstream task supervision. This formulation treats the entire perception process as a black box, often lacking explicit 3D geometric understanding and interpretability, leading to suboptimal performance. In this paper, we claim that an explicit 3D representation matters for accurate BEV perception, and we propose Splat2BEV, a Gaussian Splatting-assisted framework for BEV tasks. Splat2BEV aims to learn BEV feature representations that are both semantically rich and geometrically precise. We first pre-train a Gaussian generator that explicitly reconstructs 3D scenes from multi-view inputs, enabling the generation of geometry-aligned feature representations. These representations are then projected into the BEV space to serve as inputs for downstream tasks. Extensive experiments on nuScenes and argoverse dataset demonstrate that Splat2BEV achieves state-of-the-art performance and validate the effectiveness of incorporating explicit 3D reconstruction into BEV perception.