Range-Aware Attention Network for LiDAR-based 3D Object Detection with Auxiliary Point Density Level Estimation

TL;DR

This paper introduces RAANet, a range-aware attention network that improves LiDAR-based 3D object detection by addressing point sparsity at greater distances, achieving state-of-the-art results with real-time performance.

Contribution

The paper proposes a novel lightweight RAA convolution and an auxiliary point density estimation loss to enhance BEV feature extraction and detection accuracy for distant and occluded objects.

Findings

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

Achieves real-time inference at 16 Hz (full) and 22 Hz (lite).

Effectively improves detection of distant and occluded objects.

Abstract

3D object detection from LiDAR data for autonomous driving has been making remarkable strides in recent years. Among the state-of-the-art methodologies, encoding point clouds into a bird's eye view (BEV) has been demonstrated to be both effective and efficient. Different from perspective views, BEV preserves rich spatial and distance information between objects. Yet, while farther objects of the same type do not appear smaller in the BEV, they contain sparser point cloud features. This fact weakens BEV feature extraction using shared-weight convolutional neural networks (CNNs). In order to address this challenge, we propose Range-Aware Attention Network (RAANet), which extracts effective BEV features and generates superior 3D object detection outputs. The range-aware attention (RAA) convolutions significantly improve feature extraction for near as well as far objects. Moreover, we propose a novel auxiliary loss for point density estimation to further enhance the detection accuracy of RAANet for occluded objects. It is worth to note that our proposed RAA convolution is lightweight and compatible to be integrated into any CNN architecture used for detection from a BEV. Extensive experiments on the nuScenes and KITTI datasets demonstrate that our proposed approach outperforms the state-of-the-art methods for LiDAR-based 3D object detection, with real-time inference speed of 16 Hz for the full version and 22 Hz for the lite version tested on nuScenes lidar frames. The code is publicly available at our Github repository https://github.com/erbloo/RAAN.