Saved in:
Bibliographic Details
Main Authors: Zhou, Zixiang, Ye, Dongqiangzi, Chen, Weijia, Xie, Yufei, Wang, Yu, Wang, Panqu, Foroosh, Hassan
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2303.12194
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929261763362816
author Zhou, Zixiang
Ye, Dongqiangzi
Chen, Weijia
Xie, Yufei
Wang, Yu
Wang, Panqu
Foroosh, Hassan
author_facet Zhou, Zixiang
Ye, Dongqiangzi
Chen, Weijia
Xie, Yufei
Wang, Yu
Wang, Panqu
Foroosh, Hassan
contents There is a recent trend in the LiDAR perception field towards unifying multiple tasks in a single strong network with improved performance, as opposed to using separate networks for each task. In this paper, we introduce a new LiDAR multi-task learning paradigm based on the transformer. The proposed LiDARFormer utilizes cross-space global contextual feature information and exploits cross-task synergy to boost the performance of LiDAR perception tasks across multiple large-scale datasets and benchmarks. Our novel transformer-based framework includes a cross-space transformer module that learns attentive features between the 2D dense Bird's Eye View (BEV) and 3D sparse voxel feature maps. Additionally, we propose a transformer decoder for the segmentation task to dynamically adjust the learned features by leveraging the categorical feature representations. Furthermore, we combine the segmentation and detection features in a shared transformer decoder with cross-task attention layers to enhance and integrate the object-level and class-level features. LiDARFormer is evaluated on the large-scale nuScenes and the Waymo Open datasets for both 3D detection and semantic segmentation tasks, and it outperforms all previously published methods on both tasks. Notably, LiDARFormer achieves the state-of-the-art performance of 76.4% L2 mAPH and 74.3% NDS on the challenging Waymo and nuScenes detection benchmarks for a single model LiDAR-only method.
format Preprint
id arxiv_https___arxiv_org_abs_2303_12194
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle LiDARFormer: A Unified Transformer-based Multi-task Network for LiDAR Perception
Zhou, Zixiang
Ye, Dongqiangzi
Chen, Weijia
Xie, Yufei
Wang, Yu
Wang, Panqu
Foroosh, Hassan
Computer Vision and Pattern Recognition
There is a recent trend in the LiDAR perception field towards unifying multiple tasks in a single strong network with improved performance, as opposed to using separate networks for each task. In this paper, we introduce a new LiDAR multi-task learning paradigm based on the transformer. The proposed LiDARFormer utilizes cross-space global contextual feature information and exploits cross-task synergy to boost the performance of LiDAR perception tasks across multiple large-scale datasets and benchmarks. Our novel transformer-based framework includes a cross-space transformer module that learns attentive features between the 2D dense Bird's Eye View (BEV) and 3D sparse voxel feature maps. Additionally, we propose a transformer decoder for the segmentation task to dynamically adjust the learned features by leveraging the categorical feature representations. Furthermore, we combine the segmentation and detection features in a shared transformer decoder with cross-task attention layers to enhance and integrate the object-level and class-level features. LiDARFormer is evaluated on the large-scale nuScenes and the Waymo Open datasets for both 3D detection and semantic segmentation tasks, and it outperforms all previously published methods on both tasks. Notably, LiDARFormer achieves the state-of-the-art performance of 76.4% L2 mAPH and 74.3% NDS on the challenging Waymo and nuScenes detection benchmarks for a single model LiDAR-only method.
title LiDARFormer: A Unified Transformer-based Multi-task Network for LiDAR Perception
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2303.12194