- Browse by Author
Browsing by Author "Li, Jingjing"
Now showing 1 - 5 of 5
Results Per Page
Sort Options
Item Cycle-consistent Conditional Adversarial Transfer Networks(ACM, 2019-10) Li, Jingjing; Chen, Erpeng; Ding, Zhengming; Zhu, Lei; Lu, Ke; Huang, Zi; Computer Information and Graphics Technology, School of Engineering and TechnologyDomain adaptation investigates the problem of cross-domain knowledge transfer where the labeled source domain and unlabeled target domain have distinctive data distributions. Recently, adversarial training have been successfully applied to domain adaptation and achieved state-of-the-art performance. However, there is still a fatal weakness existing in current adversarial models which is raised from the equilibrium challenge of adversarial training. Specifically, although most of existing methods are able to confuse the domain discriminator, they cannot guarantee that the source domain and target domain are sufficiently similar. In this paper, we propose a novel approach named cycle-consistent conditional adversarial transfer networks (3CATN) to handle this issue. Our approach takes care of the domain alignment by leveraging adversarial training. Specifically, we condition the adversarial networks with the cross-covariance of learned features and classifier predictions to capture the multimodal structures of data distributions. However, since the classifier predictions are not certainty information, a strong condition with the predictions is risky when the predictions are not accurate. We, therefore, further propose that the truly domain-invariant features should be able to be translated from one domain to the other. To this end, we introduce two feature translation losses and one cycle-consistent loss into the conditional adversarial domain adaptation networks. Extensive experiments on both classical and large-scale datasets verify that our model is able to outperform previous state-of-the-arts with significant improvements.Item Leveraging the Invariant Side of Generative Zero-Shot Learning(IEEE, 2019) Li, Jingjing; Jing, Mengmeng; Lu, Ke; Ding, Zhengming; Zhu, Lei; Huang, Zi; Electrical and Computer Engineering, School of Engineering and TechnologyConventional zero-shot learning (ZSL) methods generally learn an embedding, e.g., visual-semantic mapping, to handle the unseen visual samples via an indirect manner. In this paper, we take the advantage of generative adversarial networks (GANs) and propose a novel method, named leveraging invariant side GAN (LisGAN), which can directly generate the unseen features from random noises which are conditioned by the semantic descriptions. Specifically, we train a conditional Wasserstein GANs in which the generator synthesizes fake unseen features from noises and the discriminator distinguishes the fake from real via a minimax game. Considering that one semantic description can correspond to various synthesized visual samples, and the semantic description, figuratively, is the soul of the generated features, we introduce soul samples as the invariant side of generative zero-shot learning in this paper. A soul sample is the meta-representation of one class. It visualizes the most semantically-meaningful aspects of each sample in the same category. We regularize that each generated sample (the varying side of generative ZSL) should be close to at least one soul sample (the invariant side) which has the same class label with it. At the zero-shot recognition stage, we propose to use two classifiers, which are deployed in a cascade way, to achieve a coarse-to-fine result. Experiments on five popular benchmarks verify that our proposed approach can outperform state-of-the-art methods with significant improvements.Item Maximum Density Divergence for Domain Adaptation(IEEE, 2021) Li, Jingjing; Chen, Erpeng; Ding, Zhengming; Zhu, Lei; Lu, Ke; Shen, Heng Tao; Computer Information and Graphics Technology, School of Engineering and TechnologyUnsupervised domain adaptation addresses the problem of transferring knowledge from a well-labeled source domain to an unlabeled target domain where the two domains have distinctive data distributions. Thus, the essence of domain adaptation is to mitigate the distribution divergence between the two domains. The state-of-the-art methods practice this very idea by either conducting adversarial training or minimizing a metric which defines the distribution gaps. In this paper, we propose a new domain adaptation method named adversarial tight match (ATM) which enjoys the benefits of both adversarial training and metric learning. Specifically, at first, we propose a novel distance loss, named maximum density divergence (MDD), to quantify the distribution divergence. MDD minimizes the inter-domain divergence ("match" in ATM) and maximizes the intra-class density ("tight" in ATM). Then, to address the equilibrium challenge issue in adversarial domain adaptation, we consider leveraging the proposed MDD into adversarial domain adaptation framework. At last, we tailor the proposed MDD as a practical learning loss and report our ATM. Both empirical evaluation and theoretical analysis are reported to verify the effectiveness of the proposed method. The experimental results on four benchmarks, both classical and large-scale, show that our method is able to achieve new state-of-the-art performance on most evaluations.Item Notch signaling regulates Hey2 expression in a spatiotemporal dependent manner during cardiac morphogenesis and trabecular specification(Nature Publishing Group, 2018-02-08) Miao, Lianjie; Li, Jingjing; Li, Jun; Tian, Xueying; Lu, Yangyang; Hu, Saiyang; Shieh, David; Kanai, Ryan; Zhou, Bo-yang; Zhou, Bin; Liu, Jiandong; Firulli, Anthony B.; Martin, James F.; Singer, Harold; Zhou, Bin; Xin, Hongbo; Wu, Mingfu; Pediatrics, School of MedicineHey2 gene mutations in both humans and mice have been associated with multiple cardiac defects. However, the currently reported localization of Hey2 in the ventricular compact zone cannot explain the wide variety of cardiac defects. Furthermore, it was reported that, in contrast to other organs, Notch doesn't regulate Hey2 in the heart. To determine the expression pattern and the regulation of Hey2, we used novel methods including RNAscope and a Hey2 CreERT2 knockin line to precisely determine the spatiotemporal expression pattern and level of Hey2 during cardiac development. We found that Hey2 is expressed in the endocardial cells of the atrioventricular canal and the outflow tract, as well as at the base of trabeculae, in addition to the reported expression in the ventricular compact myocardium. By disrupting several signaling pathways that regulate trabeculation and/or compaction, we found that, in contrast to previous reports, Notch signaling and Nrg1/ErbB2 regulate Hey2 expression level in myocardium and/or endocardium, but not its expression pattern: weak expression in trabecular myocardium and strong expression in compact myocardium. Instead, we found that FGF signaling regulates the expression pattern of Hey2 in the early myocardium, and regulates the expression level of Hey2 in a Notch1 dependent manner.Item Single-Cell Lineage Tracing Reveals that Oriented Cell Division Contributes to Trabecular Morphogenesis and Regional Specification(Elsevier, 2016-04-05) Li, Jingjing; Miao, Lianjie; Shieh, David; Spiotto, Ernest; Li, Jian; Zhou, Bin; Paul, Antoni; Schwartz, Robert J.; Firulli, Anthony B.; Singer, Harold A.; Huang, Guoying; Wu, Mingfu; Department of Pediatrics, IU School of MedicineThe cardiac trabeculae are sheet-like structures extending from the myocardium that function to increase surface area. A lack of trabeculation causes embryonic lethality due to compromised cardiac function. To understand the cellular and molecular mechanisms of trabecular formation, we genetically labeled individual cardiomyocytes prior to trabeculation via the brainbow multicolor system and traced and analyzed the labeled cells during trabeculation by whole-embryo clearing and imaging. The clones derived from labeled single cells displayed four different geometric patterns that are derived from different patterns of oriented cell division (OCD) and migration. Of the four types of clones, the inner, transmural, and mixed clones contributed to trabecular cardiomyocytes. Further studies showed that perpendicular OCD is an extrinsic asymmetric cell division that putatively contributes to trabecular regional specification. Furthermore, N-Cadherin deletion in labeled clones disrupted the clonal patterns. In summary, our data demonstrate that OCD contributes to trabecular morphogenesis and specification.