Analysis of voxel-based 3D object detection methods efficiency for real-time embedded systems

TL;DR

This paper evaluates the efficiency of voxel-based 3D object detection methods for real-time embedded systems, revealing that focusing on near objects can significantly improve speed without major performance loss.

Contribution

It provides an analysis of detection performance and computational efficiency, suggesting a practical speed-up approach by concentrating on near objects in voxel-based detection.

Findings

Detection of distant small objects is limited due to sparse point clouds.

Models trained on near objects perform as well or better than those trained on all objects.

Speed can be increased by 40-60% by focusing on near objects without significant accuracy loss.

Abstract

Real-time detection of objects in the 3D scene is one of the tasks an autonomous agent needs to perform for understanding its surroundings. While recent Deep Learning-based solutions achieve satisfactory performance, their high computational cost renders their application in real-life settings in which computations need to be performed on embedded platforms intractable. In this paper, we analyze the efficiency of two popular voxel-based 3D object detection methods providing a good compromise between high performance and speed based on two aspects, their ability to detect objects located at large distances from the agent and their ability to operate in real time on embedded platforms equipped with high-performance GPUs. Our experiments show that these methods mostly fail to detect distant small objects due to the sparsity of the input point clouds at large distances. Moreover, models trained on near objects achieve similar or better performance compared to those trained on all objects in the scene. This means that the models learn object appearance representations mostly from near objects. Our findings suggest that a considerable part of the computations of existing methods is focused on locations of the scene that do not contribute with successful detection. This means that the methods can achieve a speed-up of $40$-$60\%$ by restricting operation to near objects while not sacrificing much in performance.