TransPillars: Coarse-to-Fine Aggregation for Multi-Frame 3D Object Detection
This addresses the problem of improving detection accuracy for autonomous driving and robotics by leveraging temporal information, representing an incremental advance over prior multi-frame methods.
The paper tackles 3D object detection in point cloud sequences by proposing TransPillars, a transformer-based method that aggregates temporal features from consecutive frames, achieving state-of-the-art performance compared to existing multi-frame approaches.
3D object detection using point clouds has attracted increasing attention due to its wide applications in autonomous driving and robotics. However, most existing studies focus on single point cloud frames without harnessing the temporal information in point cloud sequences. In this paper, we design TransPillars, a novel transformer-based feature aggregation technique that exploits temporal features of consecutive point cloud frames for multi-frame 3D object detection. TransPillars aggregates spatial-temporal point cloud features from two perspectives. First, it fuses voxel-level features directly from multi-frame feature maps instead of pooled instance features to preserve instance details with contextual information that are essential to accurate object localization. Second, it introduces a hierarchical coarse-to-fine strategy to fuse multi-scale features progressively to effectively capture the motion of moving objects and guide the aggregation of fine features. Besides, a variant of deformable transformer is introduced to improve the effectiveness of cross-frame feature matching. Extensive experiments show that our proposed TransPillars achieves state-of-art performance as compared to existing multi-frame detection approaches. Code will be released.