Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Guo, Mingzhe, Zhang, Zhipeng, Jing, Liping, He, Yuan, Wang, Ke, Fan, Heng
Format: Preprint
Veröffentlicht: 2024
Schlagworte:
Online-Zugang:https://arxiv.org/abs/2407.03240
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866910511860285440
author Guo, Mingzhe
Zhang, Zhipeng
Jing, Liping
He, Yuan
Wang, Ke
Fan, Heng
author_facet Guo, Mingzhe
Zhang, Zhipeng
Jing, Liping
He, Yuan
Wang, Ke
Fan, Heng
contents We propose a unified object-aware temporal learning framework for multi-view 3D detection and tracking tasks. Having observed that the efficacy of the temporal fusion strategy in recent multi-view perception methods may be weakened by distractors and background clutters in historical frames, we propose a cyclic learning mechanism to improve the robustness of multi-view representation learning. The essence is constructing a backward bridge to propagate information from model predictions (e.g., object locations and sizes) to image and BEV features, which forms a circle with regular inference. After backward refinement, the responses of target-irrelevant regions in historical frames would be suppressed, decreasing the risk of polluting future frames and improving the object awareness ability of temporal fusion. We further tailor an object-aware association strategy for tracking based on the cyclic learning model. The cyclic learning model not only provides refined features, but also delivers finer clues (e.g., scale level) for tracklet association. The proposed cycle learning method and association module together contribute a novel and unified multi-task framework. Experiments on nuScenes show that the proposed model achieves consistent performance gains over baselines of different designs (i.e., dense query-based BEVFormer, sparse query-based SparseBEV and LSS-based BEVDet4D) on both detection and tracking evaluation.
format Preprint
id arxiv_https___arxiv_org_abs_2407_03240
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Cyclic Refiner: Object-Aware Temporal Representation Learning for Multi-View 3D Detection and Tracking
Guo, Mingzhe
Zhang, Zhipeng
Jing, Liping
He, Yuan
Wang, Ke
Fan, Heng
Computer Vision and Pattern Recognition
We propose a unified object-aware temporal learning framework for multi-view 3D detection and tracking tasks. Having observed that the efficacy of the temporal fusion strategy in recent multi-view perception methods may be weakened by distractors and background clutters in historical frames, we propose a cyclic learning mechanism to improve the robustness of multi-view representation learning. The essence is constructing a backward bridge to propagate information from model predictions (e.g., object locations and sizes) to image and BEV features, which forms a circle with regular inference. After backward refinement, the responses of target-irrelevant regions in historical frames would be suppressed, decreasing the risk of polluting future frames and improving the object awareness ability of temporal fusion. We further tailor an object-aware association strategy for tracking based on the cyclic learning model. The cyclic learning model not only provides refined features, but also delivers finer clues (e.g., scale level) for tracklet association. The proposed cycle learning method and association module together contribute a novel and unified multi-task framework. Experiments on nuScenes show that the proposed model achieves consistent performance gains over baselines of different designs (i.e., dense query-based BEVFormer, sparse query-based SparseBEV and LSS-based BEVDet4D) on both detection and tracking evaluation.
title Cyclic Refiner: Object-Aware Temporal Representation Learning for Multi-View 3D Detection and Tracking
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2407.03240