TacoDepth: Towards Efficient Radar-Camera Depth Estimation with One-stage Fusion
This work addresses the need for real-time, efficient depth estimation in autonomous vehicles and robotics, offering a novel approach that is incremental in improving upon existing multi-stage frameworks.
The paper tackles the problem of inefficient and non-robust depth estimation from radar-camera fusion by proposing TacoDepth, a one-stage fusion model that improves depth accuracy by 12.8% and processing speed by 91.8% compared to prior state-of-the-art methods.
Radar-Camera depth estimation aims to predict dense and accurate metric depth by fusing input images and Radar data. Model efficiency is crucial for this task in pursuit of real-time processing on autonomous vehicles and robotic platforms. However, due to the sparsity of Radar returns, the prevailing methods adopt multi-stage frameworks with intermediate quasi-dense depth, which are time-consuming and not robust. To address these challenges, we propose TacoDepth, an efficient and accurate Radar-Camera depth estimation model with one-stage fusion. Specifically, the graph-based Radar structure extractor and the pyramid-based Radar fusion module are designed to capture and integrate the graph structures of Radar point clouds, delivering superior model efficiency and robustness without relying on the intermediate depth results. Moreover, TacoDepth can be flexible for different inference modes, providing a better balance of speed and accuracy. Extensive experiments are conducted to demonstrate the efficacy of our method. Compared with the previous state-of-the-art approach, TacoDepth improves depth accuracy and processing speed by 12.8% and 91.8%. Our work provides a new perspective on efficient Radar-Camera depth estimation.