CASIA OpenIR  > 多模态人工智能系统全国重点实验室
Multi-Prototype Guided Source-Free Domain Adaptive Object Detection for Autonomous Driving
Siqi Zhang; Lu Zhang; Guangsen Li; Pengcheng Li; Zhiyong Liu
Source Publicationhttps://ieeexplore.ieee.org/document/10336548/
2024-01
Volume9Issue:1Pages:1589-1601
Subtype期刊论文
Abstract

Source-free domain adaptive object detection (source-free DAOD) seeks to adapt a detector pre-trained on a source domain to an unlabeled target domain without requiring access to annotated source domain data. To address challenges posed by domain shifts, current source-free DAOD approaches mainly rely on the self-training paradigm, where pseudo labels are predicted and employed to fine-tune the detector on unlabeled target domain. However, these methods often encounter issues related to intra-class variation, resulting in category-specific biases and noisy pseudo labels. In response, we present an effective Multi-Prototype Guided source-free DAOD method, dubbed MPG, consisting of two key components: multi-prototype guided pseudo labeling (MPPL) and multi-prototype guided consistency regularization (MPCR) modules. In the MPPL module, we construct category-specific multiple prototypes to better represent the category with intra-class variations. Specifically, multiple prototypes with adaptive cluster centroids are introduced for each category to effectively capture the intra-class variations. Through the implementation of the proposed MPPL module, we derive more accurate pseudo labels by assessing the proximity of instance features to multiple category prototypes. In the MPCR module, we introduce multi-level consistency regularization, including prototype-based consistency and prediction consistency, which encourages the model to overlook style perturbations and learn domain-invariant representations. Extensive experiments on five public driving datasets demonstrate that MPG outperforms existing state-of-the-art methods, showcasing its effectiveness in adapting object detectors to target domains.

Language英语
Sub direction classification目标检测、跟踪与识别
planning direction of the national heavy laboratory其他
Paper associated data
Document Type期刊论文
Identifierhttp://ir.ia.ac.cn/handle/173211/57279
Collection多模态人工智能系统全国重点实验室
AffiliationState Key Laboratory of Multimodal Artificial In- telligence Systems, Institute of Automation, Chinese Academy of Sciences
First Author AffilicationInstitute of Automation, Chinese Academy of Sciences
Recommended Citation
GB/T 7714
Siqi Zhang,Lu Zhang,Guangsen Li,et al. Multi-Prototype Guided Source-Free Domain Adaptive Object Detection for Autonomous Driving[J]. https://ieeexplore.ieee.org/document/10336548/,2024,9(1):1589-1601.
APA Siqi Zhang,Lu Zhang,Guangsen Li,Pengcheng Li,&Zhiyong Liu.(2024).Multi-Prototype Guided Source-Free Domain Adaptive Object Detection for Autonomous Driving.https://ieeexplore.ieee.org/document/10336548/,9(1),1589-1601.
MLA Siqi Zhang,et al."Multi-Prototype Guided Source-Free Domain Adaptive Object Detection for Autonomous Driving".https://ieeexplore.ieee.org/document/10336548/ 9.1(2024):1589-1601.
Files in This Item: Download All
File Name/Size DocType Version Access License
Zhang 等 - 2024 - Mul(4349KB)期刊论文作者接受稿开放获取CC BY-NC-SAView Download
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[Siqi Zhang]'s Articles
[Lu Zhang]'s Articles
[Guangsen Li]'s Articles
Baidu academic
Similar articles in Baidu academic
[Siqi Zhang]'s Articles
[Lu Zhang]'s Articles
[Guangsen Li]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[Siqi Zhang]'s Articles
[Lu Zhang]'s Articles
[Guangsen Li]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: Zhang 等 - 2024 - Multi-Prototype Guided Source-Free Domain Adaptive.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.