CRAFT: Camera-Radar 3D Object Detection with Spatio-Contextual Fusion Transformer
This work addresses the problem of improving autonomous vehicle perception by enhancing camera-radar fusion for 3D object detection, offering a significant performance boost over existing methods.
The paper tackles 3D object detection by proposing a novel early fusion approach that combines camera and radar data using a spatio-contextual fusion transformer, achieving state-of-the-art results of 41.1% mAP and 52.3% NDS on the nuScenes test set, which is 8.7 and 10.8 points higher than a camera-only baseline.
Camera and radar sensors have significant advantages in cost, reliability, and maintenance compared to LiDAR. Existing fusion methods often fuse the outputs of single modalities at the result-level, called the late fusion strategy. This can benefit from using off-the-shelf single sensor detection algorithms, but late fusion cannot fully exploit the complementary properties of sensors, thus having limited performance despite the huge potential of camera-radar fusion. Here we propose a novel proposal-level early fusion approach that effectively exploits both spatial and contextual properties of camera and radar for 3D object detection. Our fusion framework first associates image proposal with radar points in the polar coordinate system to efficiently handle the discrepancy between the coordinate system and spatial properties. Using this as a first stage, following consecutive cross-attention based feature fusion layers adaptively exchange spatio-contextual information between camera and radar, leading to a robust and attentive fusion. Our camera-radar fusion approach achieves the state-of-the-art 41.1% mAP and 52.3% NDS on the nuScenes test set, which is 8.7 and 10.8 points higher than the camera-only baseline, as well as yielding competitive performance on the LiDAR method.