CVFNet: Real-time 3D Object Detection by Learning Cross View Features

TL;DR

CVFNet is a real-time 3D object detection framework that efficiently fuses multi-view features from LiDAR data, achieving high accuracy and speed suitable for time-critical applications.

Contribution

The paper introduces a novel Point-Range feature fusion module and Slice Pillar design for efficient, accurate 3D detection from LiDAR data in real-time.

Findings

Achieves state-of-the-art accuracy on KITTI and NuScenes benchmarks.

Operates in real-time with high computational efficiency.

Effectively balances detection accuracy and speed.

Abstract

In recent years 3D object detection from LiDAR point clouds has made great progress thanks to the development of deep learning technologies. Although voxel or point based methods are popular in 3D object detection, they usually involve time-consuming operations such as 3D convolutions on voxels or ball query among points, making the resulting network inappropriate for time critical applications. On the other hand, 2D view-based methods feature high computing efficiency while usually obtaining inferior performance than the voxel or point based methods. In this work, we present a real-time view-based single stage 3D object detector, namely CVFNet to fulfill this task. To strengthen the cross-view feature learning under the condition of demanding efficiency, our framework extracts the features of different views and fuses them in an efficient progressive way. We first propose a novel Point-Range feature fusion module that deeply integrates point and range view features in multiple stages. Then, a special Slice Pillar is designed to well maintain the 3D geometry when transforming the obtained deep point-view features into bird's eye view. To better balance the ratio of samples, a sparse pillar detection head is presented to focus the detection on the nonempty grids. We conduct experiments on the popular KITTI and NuScenes benchmark, and state-of-the-art performances are achieved in terms of both accuracy and speed.