Contrastive Learning for Automotive mmWave Radar Detection Points Based Instance Segmentation

TL;DR

This paper introduces a contrastive learning method for radar point-based instance segmentation in automotive applications, reducing the reliance on high-quality annotations and maintaining performance with limited labeled data.

Contribution

The proposed contrastive learning approach effectively segments radar detection points with minimal supervision, utilizing pseudo labels and combined training steps for improved performance.

Findings

Achieves comparable performance with limited ground-truth data

Utilizes pseudo labels to enhance training

Flexible training settings with merged steps

Abstract

The automotive mmWave radar plays a key role in advanced driver assistance systems (ADAS) and autonomous driving. Deep learning-based instance segmentation enables real-time object identification from the radar detection points. In the conventional training process, accurate annotation is the key. However, high-quality annotations of radar detection points are challenging to achieve due to their ambiguity and sparsity. To address this issue, we propose a contrastive learning approach for implementing radar detection points-based instance segmentation. We define the positive and negative samples according to the ground-truth label, apply the contrastive loss to train the model first, and then perform fine-tuning for the following downstream task. In addition, these two steps can be merged into one, and pseudo labels can be generated for the unlabeled data to improve the performance further. Thus, there are four different training settings for our method. Experiments show that when the ground-truth information is only available for a small proportion of the training data, our method still achieves a comparable performance to the approach trained in a supervised manner with 100% ground-truth information.