DQFormer: Towards Unified LiDAR Panoptic Segmentation with Decoupled Queries

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Yu, Mei, Jianbiao, Liu, Liang, Du, Siliang, Xiao, Yilin, Ra, Jongwon, Liu, Yong, Xu, Xiao, Wu, Huifeng
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914927818571776
author Yang, Yu
Mei, Jianbiao
Liu, Liang
Du, Siliang
Xiao, Yilin
Ra, Jongwon
Liu, Yong
Xu, Xiao
Wu, Huifeng
author_facet Yang, Yu
Mei, Jianbiao
Liu, Liang
Du, Siliang
Xiao, Yilin
Ra, Jongwon
Liu, Yong
Xu, Xiao
Wu, Huifeng
contents LiDAR panoptic segmentation, which jointly performs instance and semantic segmentation for things and stuff classes, plays a fundamental role in LiDAR perception tasks. While most existing methods explicitly separate these two segmentation tasks and utilize different branches (i.e., semantic and instance branches), some recent methods have embraced the query-based paradigm to unify LiDAR panoptic segmentation. However, the distinct spatial distribution and inherent characteristics of objects(things) and their surroundings(stuff) in 3D scenes lead to challenges, including the mutual competition of things/stuff and the ambiguity of classification/segmentation. In this paper, we propose decoupling things/stuff queries according to their intrinsic properties for individual decoding and disentangling classification/segmentation to mitigate ambiguity. To this end, we propose a novel framework dubbed DQFormer to implement semantic and instance segmentation in a unified workflow. Specifically, we design a decoupled query generator to propose informative queries with semantics by localizing things/stuff positions and fusing multi-level BEV embeddings. Moreover, a query-oriented mask decoder is introduced to decode corresponding segmentation masks by performing masked cross-attention between queries and mask embeddings. Finally, the decoded masks are combined with the semantics of the queries to produce panoptic results. Extensive experiments on nuScenes and SemanticKITTI datasets demonstrate the superiority of our DQFormer framework.
format Preprint
id arxiv_https___arxiv_org_abs_2408_15813
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle DQFormer: Towards Unified LiDAR Panoptic Segmentation with Decoupled Queries
Yang, Yu
Mei, Jianbiao
Liu, Liang
Du, Siliang
Xiao, Yilin
Ra, Jongwon
Liu, Yong
Xu, Xiao
Wu, Huifeng
Computer Vision and Pattern Recognition
LiDAR panoptic segmentation, which jointly performs instance and semantic segmentation for things and stuff classes, plays a fundamental role in LiDAR perception tasks. While most existing methods explicitly separate these two segmentation tasks and utilize different branches (i.e., semantic and instance branches), some recent methods have embraced the query-based paradigm to unify LiDAR panoptic segmentation. However, the distinct spatial distribution and inherent characteristics of objects(things) and their surroundings(stuff) in 3D scenes lead to challenges, including the mutual competition of things/stuff and the ambiguity of classification/segmentation. In this paper, we propose decoupling things/stuff queries according to their intrinsic properties for individual decoding and disentangling classification/segmentation to mitigate ambiguity. To this end, we propose a novel framework dubbed DQFormer to implement semantic and instance segmentation in a unified workflow. Specifically, we design a decoupled query generator to propose informative queries with semantics by localizing things/stuff positions and fusing multi-level BEV embeddings. Moreover, a query-oriented mask decoder is introduced to decode corresponding segmentation masks by performing masked cross-attention between queries and mask embeddings. Finally, the decoded masks are combined with the semantics of the queries to produce panoptic results. Extensive experiments on nuScenes and SemanticKITTI datasets demonstrate the superiority of our DQFormer framework.
title DQFormer: Towards Unified LiDAR Panoptic Segmentation with Decoupled Queries
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2408.15813