Voxel-FPN: multi-scale voxel feature aggregation in 3D object detection from point clouds
This addresses object detection for autonomous driving systems, but it appears incremental as it builds on existing voxel-based methods with multi-scale feature fusion.
The paper tackles 3D object detection from LIDAR point clouds by proposing Voxel-FPN, a one-stage detector that aggregates multi-scale voxel features, achieving good performance in speed and accuracy on the KITTI-3D benchmark.
Object detection in point cloud data is one of the key components in computer vision systems, especially for autonomous driving applications. In this work, we present Voxel-FPN, a novel one-stage 3D object detector that utilizes raw data from LIDAR sensors only. The core framework consists of an encoder network and a corresponding decoder followed by a region proposal network. Encoder extracts multi-scale voxel information in a bottom-up manner while decoder fuses multiple feature maps from various scales in a top-down way. Extensive experiments show that the proposed method has better performance on extracting features from point data and demonstrates its superiority over some baselines on the challenging KITTI-3D benchmark, obtaining good performance on both speed and accuracy in real-world scenarios.