*: Improving the 3D detector by introducing Voxel2Pillar feature encoding and extracting multi-scale features

TL;DR

This paper introduces a novel pillar-based 3D detection scheme with Voxel2Pillar encoding and a multi-scale sparse backbone, significantly improving detection accuracy for autonomous driving applications.

Contribution

It proposes Voxel2Pillar feature encoding and a multi-scale sparse backbone, enhancing feature richness and detection performance without large convolution kernels.

Findings

Improved detection accuracy on Waymo dataset for vehicles, pedestrians, and cyclists.

Effective multi-scale feature extraction without large convolution kernels.

Ablation studies confirm the contribution of each module.

Abstract

The multi-line LiDAR is widely used in autonomous vehicles, so point cloud-based 3D detectors are essential for autonomous driving. Extracting rich multi-scale features is crucial for point cloud-based 3D detectors in autonomous driving due to significant differences in the size of different types of objects. However, because of the real-time requirements, large-size convolution kernels are rarely used to extract large-scale features in the backbone. Current 3D detectors commonly use feature pyramid networks to obtain large-scale features; however, some objects containing fewer point clouds are further lost during down-sampling, resulting in degraded performance. Since pillar-based schemes require much less computation than voxel-based schemes, they are more suitable for constructing real-time 3D detectors. Hence, we propose the *, a pillar-based scheme. We redesigned the feature encoding, the backbone, and the neck of the 3D detector. We propose the Voxel2Pillar feature encoding, which uses a sparse convolution constructor to construct pillars with richer point cloud features, especially height features. The Voxel2Pillar adds more learnable parameters to the feature encoding, enabling the initial pillars to have higher performance ability. We extract multi-scale and large-scale features in the proposed fully sparse backbone, which does not utilize large-size convolutional kernels; the backbone consists of the proposed multi-scale feature extraction module. The neck consists of the proposed sparse ConvNeXt, whose simple structure significantly improves the performance. We validate the effectiveness of the proposed * on the Waymo Open Dataset, and the object detection accuracy for vehicles, pedestrians, and cyclists is improved. We also verify the effectiveness of each proposed module in detail through ablation studies.