DifFUSER: Diffusion Model for Robust Multi-Sensor Fusion in 3D Object Detection and BEV Segmentation

TL;DR

DifFUSER introduces a diffusion model-based multi-sensor fusion approach for 3D object detection and BEV segmentation, enhancing robustness and performance, especially under sensor failure conditions.

Contribution

The paper presents DifFUSER, a novel diffusion model framework with hierarchical architecture and training paradigms that improve multi-sensor fusion robustness and accuracy.

Findings

Achieves 70.04% mIOU in BEV segmentation on Nuscenes.

Outperforms existing methods in sensor failure scenarios.

Competitive with transformer-based fusion techniques.

Abstract

Diffusion models have recently gained prominence as powerful deep generative models, demonstrating unmatched performance across various domains. However, their potential in multi-sensor fusion remains largely unexplored. In this work, we introduce DifFUSER, a novel approach that leverages diffusion models for multi-modal fusion in 3D object detection and BEV map segmentation. Benefiting from the inherent denoising property of diffusion, DifFUSER is able to refine or even synthesize sensor features in case of sensor malfunction, thereby improving the quality of the fused output. In terms of architecture, our DifFUSER blocks are chained together in a hierarchical BiFPN fashion, termed cMini-BiFPN, offering an alternative architecture for latent diffusion. We further introduce a Gated Self-conditioned Modulated (GSM) latent diffusion module together with a Progressive Sensor Dropout Training (PSDT) paradigm, designed to add stronger conditioning to the diffusion process and robustness to sensor failures. Our extensive evaluations on the Nuscenes dataset reveal that DifFUSER not only achieves state-of-the-art performance with a 70.04% mIOU in BEV map segmentation tasks but also competes effectively with leading transformer-based fusion techniques in 3D object detection.