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

5 Dec 2020  ·  Wu Zheng, Weiliang Tang, Sijin Chen, Li Jiang, Chi-Wing Fu ·

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

PDF Abstract

Datasets


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Birds Eye View Object Detection KITTI Cars Easy CIA-SSD AP 93.74 % # 3
3D Object Detection KITTI Cars Easy CIA-SSD AP 89.59% # 9
Birds Eye View Object Detection KITTI Cars Hard CIA-SSD AP 82.39 % # 4
3D Object Detection KITTI Cars Hard CIA-SSD AP 72.87 # 12
Birds Eye View Object Detection KITTI Cars Moderate CIA-SSD AP 89.84 % # 3
3D Object Detection KITTI Cars Moderate CIA-SSD AP 80.28% # 13

Methods


No methods listed for this paper. Add relevant methods here