CIA-SSD: Confident IoU-Aware Single-Stage Object Detector From Point Cloud

Existing single-stage detectors for locating objects in point clouds often treat object localization and category classification as separate tasks, so the accuracy confidence may not well align. To address this issue, we present a new detector named Confident IoU-Aware Single-Stage Detector (CIA-SSD). First, design lightweight Spatial-Semantic Feature Aggregation module to adaptively fuse high-level abstract semantic features low-level spatial accurate predictions of bounding boxes confidence. Also, predicted is further rectified with our designed IoU-aware rectification make more consistent accuracy. Based on confidence, formulate Distance-variant IoU-weighted NMS obtain smoother regressions avoid redundant predictions. We experiment CIA-SSD 3D car detection KITTI test set show that it attains top performance terms official ranking metric (moderate AP 80.28%) above 32 FPS inference speed, outperforming all prior detectors. The code available at https://github.com/Vegeta2020/CIA-SSD.