IS-Fusion: Instance-Scene Collaborative Fusion for Multimodal 3D Object Detection

TL;DR

IS-Fusion introduces a novel multimodal fusion framework that jointly captures instance- and scene-level context to improve 3D object detection in autonomous driving, outperforming existing methods on the nuScenes benchmark.

Contribution

It explicitly incorporates instance-level multimodal information through hierarchical scene and instance-guided fusion modules, enhancing 3D perception accuracy.

Findings

Outperforms all published multimodal methods on nuScenes

Effectively captures multi-granularity scene context

Enhances instance-centric 3D object detection

Abstract

Bird's eye view (BEV) representation has emerged as a dominant solution for describing 3D space in autonomous driving scenarios. However, objects in the BEV representation typically exhibit small sizes, and the associated point cloud context is inherently sparse, which leads to great challenges for reliable 3D perception. In this paper, we propose IS-Fusion, an innovative multimodal fusion framework that jointly captures the Instance- and Scene-level contextual information. IS-Fusion essentially differs from existing approaches that only focus on the BEV scene-level fusion by explicitly incorporating instance-level multimodal information, thus facilitating the instance-centric tasks like 3D object detection. It comprises a Hierarchical Scene Fusion (HSF) module and an Instance-Guided Fusion (IGF) module. HSF applies Point-to-Grid and Grid-to-Region transformers to capture the multimodal scene context at different granularities. IGF mines instance candidates, explores their relationships, and aggregates the local multimodal context for each instance. These instances then serve as guidance to enhance the scene feature and yield an instance-aware BEV representation. On the challenging nuScenes benchmark, IS-Fusion outperforms all the published multimodal works to date. Code is available at: https://github.com/yinjunbo/IS-Fusion.