Saved in:
Bibliographic Details
Main Authors: Wilson, Benjamin, Mitchell, Nicholas Autio, Pontes, Jhony Kaesemodel, Hays, James
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.16789
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916336837328896
author Wilson, Benjamin
Mitchell, Nicholas Autio
Pontes, Jhony Kaesemodel
Hays, James
author_facet Wilson, Benjamin
Mitchell, Nicholas Autio
Pontes, Jhony Kaesemodel
Hays, James
contents Lidar-based perception pipelines rely on 3D object detection models to interpret complex scenes. While multiple representations for lidar exist, the range-view is enticing since it losslessly encodes the entire lidar sensor output. In this work, we achieve state-of-the-art amongst range-view 3D object detection models without using multiple techniques proposed in past range-view literature. We explore range-view 3D object detection across two modern datasets with substantially different properties: Argoverse 2 and Waymo Open. Our investigation reveals key insights: (1) input feature dimensionality significantly influences the overall performance, (2) surprisingly, employing a classification loss grounded in 3D spatial proximity works as well or better compared to more elaborate IoU-based losses, and (3) addressing non-uniform lidar density via a straightforward range subsampling technique outperforms existing multi-resolution, range-conditioned networks. Our experiments reveal that techniques proposed in recent range-view literature are not needed to achieve state-of-the-art performance. Combining the above findings, we establish a new state-of-the-art model for range-view 3D object detection -- improving AP by 2.2% on the Waymo Open dataset while maintaining a runtime of 10 Hz. We establish the first range-view model on the Argoverse 2 dataset and outperform strong voxel-based baselines. All models are multi-class and open-source. Code is available at https://github.com/benjaminrwilson/range-view-3d-detection.
format Preprint
id arxiv_https___arxiv_org_abs_2407_16789
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle What Matters in Range View 3D Object Detection
Wilson, Benjamin
Mitchell, Nicholas Autio
Pontes, Jhony Kaesemodel
Hays, James
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Lidar-based perception pipelines rely on 3D object detection models to interpret complex scenes. While multiple representations for lidar exist, the range-view is enticing since it losslessly encodes the entire lidar sensor output. In this work, we achieve state-of-the-art amongst range-view 3D object detection models without using multiple techniques proposed in past range-view literature. We explore range-view 3D object detection across two modern datasets with substantially different properties: Argoverse 2 and Waymo Open. Our investigation reveals key insights: (1) input feature dimensionality significantly influences the overall performance, (2) surprisingly, employing a classification loss grounded in 3D spatial proximity works as well or better compared to more elaborate IoU-based losses, and (3) addressing non-uniform lidar density via a straightforward range subsampling technique outperforms existing multi-resolution, range-conditioned networks. Our experiments reveal that techniques proposed in recent range-view literature are not needed to achieve state-of-the-art performance. Combining the above findings, we establish a new state-of-the-art model for range-view 3D object detection -- improving AP by 2.2% on the Waymo Open dataset while maintaining a runtime of 10 Hz. We establish the first range-view model on the Argoverse 2 dataset and outperform strong voxel-based baselines. All models are multi-class and open-source. Code is available at https://github.com/benjaminrwilson/range-view-3d-detection.
title What Matters in Range View 3D Object Detection
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2407.16789