Instance-Guided Radar Depth Estimation for 3D Object Detection

TL;DR

This paper introduces InstaRadar, a novel Radar-guided depth estimation method that enhances monocular 3D detection by improving Radar density and semantic alignment, leading to better depth features and detection accuracy.

Contribution

The paper presents InstaRadar, an instance segmentation-guided expansion technique, and integrates a pre-trained depth model into BEVDepth, advancing Radar-camera fusion for 3D detection.

Findings

InstaRadar achieves state-of-the-art Radar-guided depth estimation.

Integration of InstaRadar improves 3D detection performance over baseline models.

The framework demonstrates steady gains, highlighting the effectiveness of explicit depth supervision.

Abstract

Accurate depth estimation is fundamental to 3D perception in autonomous driving, supporting tasks such as detection, tracking, and motion planning. However, monocular camera-based 3D detection suffers from depth ambiguity and reduced robustness under challenging conditions. Radar provides complementary advantages such as resilience to poor lighting and adverse weather, but its sparsity and low resolution limit its direct use in detection frameworks. This motivates the need for effective Radar-camera fusion with improved preprocessing and depth estimation strategies. We propose an end-to-end framework that enhances monocular 3D object detection through two key components. First, we introduce InstaRadar, an instance segmentation-guided expansion method that leverages pre-trained segmentation masks to enhance Radar density and semantic alignment, producing a more structured representation. InstaRadar achieves state-of-the-art results in Radar-guided depth estimation, showing its effectiveness in generating high-quality depth features. Second, we integrate the pre-trained RCDPT into the BEVDepth framework as a replacement for its depth module. With InstaRadar-enhanced inputs, the RCDPT integration consistently improves 3D detection performance. Overall, these components yield steady gains over the baseline BEVDepth model, demonstrating the effectiveness of InstaRadar and the advantage of explicit depth supervision in 3D object detection. Although the framework lags behind Radar-camera fusion models that directly extract BEV features, since Radar serves only as guidance rather than an independent feature stream, this limitation highlights potential for improvement. Future work will extend InstaRadar to point cloud-like representations and integrate a dedicated Radar branch with temporal cues for enhanced BEV fusion.