no code implementations • 27 Oct 2023 • Jiesi Hu, Yanwu Yang, Xutao Guo, Jinghua Wang, Ting Ma
Source-free domain adaptation (SFDA) aims to adapt models trained on a labeled source domain to an unlabeled target domain without the access to source data.
no code implementations • CVPR 2023 • Yabo Liu, Jinghua Wang, Chao Huang, YaoWei Wang, Yong Xu
To overcome these problems, we propose a cross-modality graph reasoning adaptation (CIGAR) method to take advantage of both visual and linguistic knowledge.
no code implementations • 17 Sep 2022 • Zhanyuan Yang, Jinghua Wang, Yingying Zhu
In the meta-training stage, we propose a cross-view episodic training mechanism to perform the nearest centroid classification on two different views of the same episode and adopt a distance-scaled contrastive loss based on them.
no code implementations • 3 Sep 2022 • Tianjiao Li, Lin Geng Foo, Qiuhong Ke, Hossein Rahmani, Anran Wang, Jinghua Wang, Jun Liu
We design a novel Dynamic Spatio-Temporal Specialization (DSTS) module, which consists of specialized neurons that are only activated for a subset of samples that are highly similar.
no code implementations • 12 Feb 2021 • Kanya Mo, Shen Zheng, Xiwei Wang, Jinghua Wang, Klaus-Dieter Schewe
The fully connected (FC) layer, one of the most fundamental modules in artificial neural networks (ANN), is often considered difficult and inefficient to train due to issues including the risk of overfitting caused by its large amount of parameters.
no code implementations • 11 Sep 2020 • Jinghua Wang, Jianmin Jiang
In comparison with the existing work in similar areas, our objective function has two learning targets, which are created to be jointly optimized to achieve the best possible unsupervised learning and knowledge discovery from unlabeled data sets.
no code implementations • ECCV 2020 • Jinghua Wang, Jianmin Jiang
With the hypothesis that the shift between a given pair of domains is shared across tasks, we propose a new method for ZSDA by transferring domain shift from an irrelevant task (IrT) to the task of interest (ToI).
no code implementations • 11 Sep 2020 • Jinghua Wang, Adrian Hilton, Jianmin Jiang
This paper proposes a new network structure for unsupervised deep representation learning based on spectral analysis, which is a popular technique with solid theory foundations.
no code implementations • ICCV 2019 • Jinghua Wang, Jianmin Jiang
To train CoCoGAN in the absence of target-domain data for RT, we propose a new supervisory signal, i. e. the alignment between representations across tasks.
no code implementations • 11 Sep 2020 • Jinghua Wang, Jianmin Jiang
In this paper, we propose a deep spectral analysis network for unsupervised representation learning and image clustering.
no code implementations • 3 Aug 2016 • Jinghua Wang, Zhenhua Wang, DaCheng Tao, Simon See, Gang Wang
In this paper, we tackle the problem of RGB-D semantic segmentation of indoor images.
no code implementations • 17 Nov 2015 • Jinghua Wang, Gang Wang
Recognizing actions from still images is popularly studied recently.
no code implementations • 17 Nov 2015 • Jinghua Wang, Abrar Abdul Nabi, Gang Wang, Chengde Wan, Tian-Tsong Ng
Given attributes as representations, we propose to learn a ranking SPN (sum product networks) to rank pairs of fashion images.