CenterNet3D: An Anchor Free Object Detector for Point Cloud

TL;DR

CenterNet3D introduces an anchor-free, efficient 3D object detection method from point clouds, utilizing keypoint estimation and boundary attention to outperform existing anchor-based methods in speed and accuracy.

Contribution

The paper proposes a novel anchor-free 3D detection network that models objects via center points, incorporating corner attention and confidence alignment for improved performance.

Findings

Outperforms state-of-the-art anchor-based methods on KITTI and nuScenes datasets.

Achieves 20 FPS inference speed, balancing speed and accuracy.

Eliminates the need for post-processing like non-maximum suppression.

Abstract

Accurate and fast 3D object detection from point clouds is a key task in autonomous driving. Existing one-stage 3D object detection methods can achieve real-time performance, however, they are dominated by anchor-based detectors which are inefficient and require additional post-processing. In this paper, we eliminate anchors and model an object as a single point--the center point of its bounding box. Based on the center point, we propose an anchor-free CenterNet3D network that performs 3D object detection without anchors. Our CenterNet3D uses keypoint estimation to find center points and directly regresses 3D bounding boxes. However, because inherent sparsity of point clouds, 3D object center points are likely to be in empty space which makes it difficult to estimate accurate boundaries. To solve this issue, we propose an extra corner attention module to enforce the CNN backbone to pay more attention to object boundaries. Besides, considering that one-stage detectors suffer from the discordance between the predicted bounding boxes and corresponding classification confidences, we develop an efficient keypoint-sensitive warping operation to align the confidences to the predicted bounding boxes. Our proposed CenterNet3D is non-maximum suppression free which makes it more efficient and simpler. We evaluate CenterNet3D on the widely used KITTI dataset and more challenging nuScenes dataset. Our method outperforms all state-of-the-art anchor-based one-stage methods and has comparable performance to two-stage methods as well. It has an inference speed of 20 FPS and achieves the best speed and accuracy trade-off. Our source code will be released at https://github.com/wangguojun2018/CenterNet3d.