Ego-Motion Estimation and Dynamic Motion Separation from 3D Point Clouds for Accumulating Data and Improving 3D Object Detection

TL;DR

This paper investigates how ego-motion estimation and dynamic motion correction can enhance the accumulation of radar point clouds, thereby improving 3D object detection in automotive radar systems.

Contribution

It introduces a learning-based dynamic motion estimation method and analyzes the limitations of accumulating radar point clouds for better object detection.

Findings

Improved object detection performance with ego-motion and motion correction

Analysis of radar point cloud accumulation limitations

Effectiveness of learning-based motion estimation

Abstract

New 3+1D high-resolution radar sensors are gaining importance for 3D object detection in the automotive domain due to their relative affordability and improved detection compared to classic low-resolution radar sensors. One limitation of high-resolution radar sensors, compared to lidar sensors, is the sparsity of the generated point cloud. This sparsity could be partially overcome by accumulating radar point clouds of subsequent time steps. This contribution analyzes limitations of accumulating radar point clouds on the View-of-Delft dataset. By employing different ego-motion estimation approaches, the dataset's inherent constraints, and possible solutions are analyzed. Additionally, a learning-based instance motion estimation approach is deployed to investigate the influence of dynamic motion on the accumulated point cloud for object detection. Experiments document an improved object detection performance by applying an ego-motion estimation and dynamic motion correction approach.