3D object detectors usually rely on hand-crafted proxies, e.g., anchors or centers, and translate well-studied 2D frameworks to 3D. Thus, sparse voxel features need to be densified and processed by dense prediction heads, which inevitably costs extra computation. In this paper, we instead propose VoxelNext for fully sparse 3D object detection. Our core insight is to predict objects directly based on sparse voxel features, without relying on hand-crafted proxies. Our strong sparse convolutional network VoxelNeXt detects and tracks 3D objects through voxel features entirely. It is an elegant and efficient framework, with no need for sparse-to-dense conversion or NMS post-processing. Our method achieves a better speed-accuracy trade-off than other mainframe detectors on the nuScenes dataset. For the first time, we show that a fully sparse voxel-based representation works decently for LIDAR 3D object detection and tracking. Extensive experiments on nuScenes, Waymo, and Argoverse2 benchmarks validate the effectiveness of our approach. Without bells and whistles, our model outperforms all existing LIDAR methods on the nuScenes tracking test benchmark.

翻译：3D目标检测器通常依赖手工设计的代理（如锚点或中心点），并将成熟的2D框架迁移至3D领域。因此，稀疏体素特征需要经过稠密化处理并由密集预测头进行运算，这不可避免地增加了额外计算成本。本文提出VoxelNext以实现全稀疏3D目标检测，核心思路是直接基于稀疏体素特征预测目标，无需依赖手工设计的代理。我们构建了强大的稀疏卷积网络VoxelNeXt，完全通过体素特征完成3D目标的检测与跟踪。该框架简洁高效，无需进行稀疏到稠密的转换或NMS后处理。在nuScenes数据集上，我们的方法实现了优于其他主流检测器的速度-精度权衡。我们首次证明，全稀疏体素表征能够有效支持LIDAR 3D目标检测与跟踪。在nuScenes、Waymo和Argoverse2基准上的大量实验验证了方法的有效性。无需额外技巧，我们的模型在nuScenes跟踪测试基准上超越了所有现有LIDAR方法。