A Simple Baseline for BEV Perception Without LiDAR

06/16/2022
by   Adam W. Harley, et al.
0

Building 3D perception systems for autonomous vehicles that do not rely on LiDAR is a critical research problem because of the high expense of LiDAR systems compared to cameras and other sensors. Current methods use multi-view RGB data collected from cameras around the vehicle and neurally "lift" features from the perspective images to the 2D ground plane, yielding a "bird's eye view" (BEV) feature representation of the 3D space around the vehicle. Recent research focuses on the way the features are lifted from images to the BEV plane. We instead propose a simple baseline model, where the "lifting" step simply averages features from all projected image locations, and find that it outperforms the current state-of-the-art in BEV vehicle segmentation. Our ablations show that batch size, data augmentation, and input resolution play a large part in performance. Additionally, we reconsider the utility of radar input, which has previously been either ignored or found non-helpful by recent works. With a simple RGB-radar fusion module, we obtain a sizable boost in performance, approaching the accuracy of a LiDAR-enabled system.

READ FULL TEXT

page 3

page 7

research
09/26/2022

Center Feature Fusion: Selective Multi-Sensor Fusion of Center-based Objects

Leveraging multi-modal fusion, especially between camera and LiDAR, has ...
research
05/25/2023

RC-BEVFusion: A Plug-In Module for Radar-Camera Bird's Eye View Feature Fusion

Radars and cameras belong to the most frequently used sensors for advanc...
research
11/14/2022

LAPTNet: LiDAR-Aided Perspective Transform Network

Semantic grids are a useful representation of the environment around a r...
research
04/17/2020

Lidar for Autonomous Driving: The principles, challenges, and trends for automotive lidar and perception systems

Autonomous vehicles rely on their perception systems to acquire informat...
research
10/06/2019

High-Resolution Traffic Sensing with Autonomous Vehicles

The last decades have witnessed the breakthrough of autonomous vehicles ...
research
05/03/2018

BirdNet: a 3D Object Detection Framework from LiDAR information

Understanding driving situations regardless the conditions of the traffi...
research
01/29/2023

Fast-BEV: A Fast and Strong Bird's-Eye View Perception Baseline

Recently, perception task based on Bird's-Eye View (BEV) representation ...

Please sign up or login with your details

Forgot password? Click here to reset