no code implementations • 6 Feb 2024 • Jinjing Zhu, Zhedong Hu, Tae-Kyun Kim, Lin Wang
Our framework incorporates two novel components: energy-based feature fusion (EB2F) and energy-based reliable fusion Assessment (RFA) modules.
no code implementations • 10 Jan 2024 • Jinjing Zhu, Yucheng Chen, Lin Wang
We then propose a Task-irrelevant data-Guided Knowledge Transfer (TGKT) module that transfers knowledge from the source model to the target model by leveraging the paired TI data.
no code implementations • 8 Dec 2023 • Jinjing Zhu, Feiyang Ye, Qiao Xiao, Pengxin Guo, Yu Zhang, Qiang Yang
Specifically, the proposed LIWUDA method constructs a weight network to assign weights to each instance based on its probability of belonging to common classes, and designs Weighted Optimal Transport (WOT) for domain alignment by leveraging instance weights.
no code implementations • 7 Oct 2023 • Jingyi Pan, Sihang Li, Yucheng Chen, Jinjing Zhu, Lin Wang
Moreover, semantic segmentation models trained on daytime datasets often face difficulties in generalizing effectively to nighttime conditions.
no code implementations • ICCV 2023 • Jinjing Zhu, Yunhao Luo, Xu Zheng, Hao Wang, Lin Wang
In this paper, we strive to answer the question "how to collaboratively learn convolutional neural network (CNN)-based and vision transformer (ViT)-based models by selecting and exchanging the reliable knowledge between them for semantic segmentation?"
no code implementations • 10 Jul 2023 • Yexin Liu, Weiming Zhang, Guoyang Zhao, Jinjing Zhu, Athanasios Vasilakos, Lin Wang
we propose the first test-time adaptation (TTA) framework, dubbed Night-TTA, to address the problems for nighttime RGBT semantic segmentation without access to the source (daytime) data during adaptation.
no code implementations • CVPR 2023 • Xu Zheng, Jinjing Zhu, Yexin Liu, Zidong Cao, Chong Fu, Lin Wang
Moreover, adversarial intra-projection training is proposed to reduce the inherent gap, between the features of the pinhole images and those of the ERP and TP images, respectively.
1 code implementation • CVPR 2023 • Jinjing Zhu, Haotian Bai, Lin Wang
We solve this problem from a game theory's perspective with the proposed model dubbed as PMTrans, which bridges source and target domains with an intermediate domain.
Ranked #1 on Unsupervised Domain Adaptation on Office-Home
no code implementations • 11 Dec 2022 • Yiqi Lin, Huabin Zheng, Huaping Zhong, Jinjing Zhu, Weijia Li, Conghui He, Lin Wang
To address these issues, we build a task-specific self-supervised pre-training framework from a data selection perspective based on a simple hypothesis that pre-training on the unlabeled samples with similar distribution to the target task can bring substantial performance gains.
2 code implementations • British Machine Vision Conference 2022 • Pengxin Guo, Jinjing Zhu, Yu Zhang
To solve this problem, we propose a Selective Partial Domain Adaptation (SPDA) method, which selects useful data for the adaptation to the target domain.
Ranked #1 on Partial Domain Adaptation on VisDA2017
1 code implementation • 21 May 2022 • Hao Ai, Zidong Cao, Jinjing Zhu, Haotian Bai, Yucheng Chen, Lin Wang
Omnidirectional image (ODI) data is captured with a 360x180 field-of-view, which is much wider than the pinhole cameras and contains richer spatial information than the conventional planar images.