Querying Autonomous Vehicle Point Clouds: Enhanced by 3D Object Counting with CounterNet

TL;DR

This paper introduces CounterNet, a heatmap-based neural network that significantly improves object counting accuracy in 3D point cloud data for autonomous vehicles, enhancing query reliability for various analytical tasks.

Contribution

The work presents CounterNet, a novel counting model tailored for 3D point clouds, with a feature partitioning and dynamic model selection strategy to handle complex traffic scenes.

Findings

CounterNet improves counting accuracy by 5% to 20%.

Enhanced counting leads to more reliable query results.

The approach effectively handles diverse object sizes in traffic scenes.

Abstract

Autonomous vehicles generate massive volumes of point cloud data, yet only a subset is relevant for specific tasks such as collision detection, traffic analysis, or congestion monitoring. Effectively querying this data is essential to enable targeted analytics. In this work, we formalize point cloud querying by defining three core query types: RETRIEVAL, COUNT, and AGGREGATION, each aligned with distinct analytical scenarios. All these queries rely heavily on accurate object counts to produce meaningful results, making precise object counting a critical component of query execution. Prior work has focused on indexing techniques for 2D video data, assuming detection models provide accurate counting information. However, when applied to 3D point cloud data, state-of-the-art detection models often fail to generate reliable object counts, leading to substantial errors in query results. To address this limitation, we propose CounterNet, a heatmap-based network designed for accurate object counting in large-scale point cloud data. Rather than focusing on accurate object localization, CounterNet detects object presence by finding object centers to improve counting accuracy. We further enhance its performance with a feature map partitioning strategy using overlapping regions, enabling better handling of both small and large objects in complex traffic scenes. To adapt to varying frame characteristics, we introduce a per-frame dynamic model selection strategy that selects the most effective configuration for each input. Evaluations on three real-world autonomous vehicle datasets show that CounterNet improves counting accuracy by 5% to 20% across object categories, resulting in more reliable query outcomes across all supported query types.