Six-Channel Image Representation for Cross-Domain Object Detection

Zhang, Tianxiao; Ma, Wenchi; Wang, Guanghui

doi:10.1007/978-3-030-87355-4_15

Cited by 5 publications

(1 citation statement)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Domain Adaptive Object Detection (DAOD) (Cai et al 2019;Chen et al 2018;Xu et al 2020b;Gu, Sun, and Xu 2020;Zhang, Ma, and Wang 2021;Csaba et al 2021) aims to address the domain shift problems in object detection task. Existing DAOD methods can be divided into two categories: feature alignment methods and self-training methods.…”

Section: Related Workmentioning

confidence: 99%

Adversarial Alignment for Source Free Object Detection

Chu,

Li,

Chen

et al. 2023

AAAI

View full text Add to dashboard Cite

Source-free object detection (SFOD) aims to transfer a detector pre-trained on a label-rich source domain to an unlabeled target domain without seeing source data. While most existing SFOD methods generate pseudo labels via a source-pretrained model to guide training, these pseudo labels usually contain high noises due to heavy domain discrepancy. In order to obtain better pseudo supervisions, we divide the target domain into source-similar and source-dissimilar parts and align them in the feature space by adversarial learning.Specifically, we design a detection variance-based criterion to divide the target domain. This criterion is motivated by a finding that larger detection variances denote higher recall and larger similarity to the source domain. Then we incorporate an adversarial module into a mean teacher framework to drive the feature spaces of these two subsets indistinguishable. Extensive experiments on multiple cross-domain object detection datasets demonstrate that our proposed method consistently outperforms the compared SFOD methods. Our implementation is available at https://github.com/ChuQiaosong.

show abstract