Class-Aware PillarMix: Can Mixed Sample Data Augmentation Enhance 3D Object Detection with Radar Point Clouds?

2025-03-04

AI Summary

This work addresses the failure of Mixed Sample Data Augmentation (MSDA) for radar point clouds—caused by sparsity, irregular sensor placement, and non-uniform angular distribution. We propose a class-aware pillar-level MixUp method. Its core innovation is the first-ever class-aware, pillar-wise independent mixing ratio mechanism, which dynamically adjusts source point sampling bias according to category-specific density (e.g., vehicles vs. pedestrians), thereby preserving geometric fidelity while enhancing sample diversity. The method integrates pillar-based feature representation, class-aware mixing, and density-adaptive sampling. Evaluated on Bosch Street and K-Radar benchmarks, it achieves significant improvements in 3D object detection accuracy, consistently outperforming existing MSDA approaches. This work establishes a new paradigm for data augmentation tailored to radar point cloud detection.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Multi-modal Vision
• Search and Optimization: Mixed Discrete/Continuous Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Due to the significant effort required for data collection and annotation in 3D perception tasks, mixed sample data augmentation (MSDA) has been widely studied to generate diverse training samples by mixing existing data. Recently, many MSDA techniques have been developed for point clouds, but they mainly target LiDAR data, leaving their application to radar point clouds largely unexplored. In this paper, we examine the feasibility of applying existing MSDA methods to radar point clouds and identify several challenges in adapting these techniques. These obstacles stem from the radar's irregular angular distribution, deviations from a single-sensor polar layout in multi-radar setups, and point sparsity. To address these issues, we propose Class-Aware PillarMix (CAPMix), a novel MSDA approach that applies MixUp at the pillar level in 3D point clouds, guided by class labels. Unlike methods that rely a single mix ratio to the entire sample, CAPMix assigns an independent ratio to each pillar, boosting sample diversity. To account for the density of different classes, we use class-specific distributions: for dense objects (e.g., large vehicles), we skew ratios to favor points from another sample, while for sparse objects (e.g., pedestrians), we sample more points from the original. This class-aware mixing retains critical details and enriches each sample with new information, ultimately generating more diverse training data. Experimental results demonstrate that our method not only significantly boosts performance but also outperforms existing MSDA approaches across two datasets (Bosch Street and K-Radar). We believe that this straightforward yet effective approach will spark further investigation into MSDA techniques for radar data.

Problem

• Explores MSDA application to radar point clouds
• Addresses challenges in radar data augmentation
• Proposes CAPMix for enhanced 3D object detection

Innovation

• Class-Aware PillarMix for radar point clouds
• Pillar-level MixUp with class-specific distributions
• Enhances 3D object detection diversity