Single image dehazing is a challenging ill-posed problem due to the severe information degeneration. However, existing deep learning based dehazing methods only adopt clear images as positive samples ...to guide the training of dehazing network while negative information is unexploited. Moreover, most of them focus on strengthening the dehazing network with an increase of depth and width, leading to a significant requirement of computation and memory. In this paper, we propose a novel contrastive regularization (CR) built upon contrastive learning to exploit both the information of hazy images and clear images as negative and positive samples, respectively. CR ensures that the restored image is pulled to closer to the clear image and pushed to far away from the hazy image in the representation space.Furthermore, considering trade-off between performance and memory storage, we develop a compact dehazing network based on autoencoder-like (AE) framework. It involves an adaptive mixup operation and a dynamic feature enhancement module, which can benefit from preserving information flow adaptively and expanding the receptive field to improve the network's transformation capability, respectively. We term our dehazing network with autoencoder and contrastive regularization as AECR-Net. The extensive experiments on synthetic and real-world datasets demonstrate that our AECR-Net surpass the state-of-the-art approaches. The code is released in https://github.com/GlassyWu/AECR-Net.
We propose a new approach to image segmentation, which exploits the advantages of both conditional random fields (CRFs) and decision trees. In the literature, the potential functions of CRFs are ...mostly defined as a linear combination of some predefined parametric models, and then, methods, such as structured support vector machines, are applied to learn those linear coefficients. We instead formulate the unary and pairwise potentials as nonparametric forests-ensembles of decision trees, and learn the ensemble parameters and the trees in a unified optimization problem within the large-margin framework. In this fashion, we easily achieve nonlinear learning of potential functions on both unary and pairwise terms in CRFs. Moreover, we learn classwise decision trees for each object that appears in the image. Experimental results on several public segmentation data sets demonstrate the power of the learned nonlinear nonparametric potentials.
In recent studies, combinations of histone deacetylases (HDACs) inhibitor with kinase inhibitor showed additive and synergistic effects. BRafV600E as an attractive target in many diseases treatments ...has been studied extensively. Herein, we present a novel design approach though incorporating the pharmacophores of BRafV600E inhibitor and HDACs inhibitor in one molecule. Several synthesized compounds exhibited distinct BRafV600E and HDAC1 inhibitory activities. The representative dual Raf/HDAC inhibitor, 7a, showed better antiproliferative activities against A549 and SK-Mel-2 in cellular assay than SAHA and sorafenib, with IC50 values of 9.11 µM and 5.40 µM, respectively. This work may lay the foundation for the further development of dual Raf/HDAC inhibitors as potential anticancer agents.
Multi-dataset Detection with Transformers Ke, Bo; Qiao, Ruizhi; Sun, Xing
International journal of computer vision,
07/2024, Letnik:
132, Številka:
7
Journal Article
Recenzirano
Learning a unified model from multiple datasets is very challenging. In this paper, we propose a multi-dataset detector using the transformer (MDT). To enhance the effectiveness of the fusion of ...multiple datasets, we propose alternative learning to suppress the noisy data. To speed up the training of big data, we use scale shifting to save computational effort. Experiments on OpenImages, COCO, and Mapillary datasets show that our approach can significantly accelerate training while improving performance on multiple datasets. In the Robust Vision Challenge 2022, our solution won 1st place on the object detection track.
Devising a representation suitable for characterizing human actions on the basis of a sequence of pose estimates generated by an RGBD sensor remains a research challenge. We here provide two insights ...into this challenge. First, we show that discriminate sequence of poses typically occur over a short time window, and thus we propose a simple-but-effective local descriptor called a trajectorylet to capture the static and kinematic information within this interval. Second, we show that state of the art recognition results can be achieved by encoding each trajectorylet using a discriminative trajectorylet detector set which is selected from a large number of candidate detectors trained through exemplar-SVMs. The action-level representation is obtained by pooling trajectorylet encodings. Evaluating on standard datasets acquired from the Kinect sensor, it is demonstrated that our method obtains superior results over existing approaches under various experimental setups.
•We design the trajectorylet that captures the static and dynamic pose information withina short interval.•We obtain discriminative patterns of action instances by encoding each trajectorylet using a discriminative trajectorylet detector set.•This method generalizes well to different parameter settings and datasets.
Classifying a visual concept merely from its associated online textual source, such as a Wikipedia article, is an attractive research topic in zero-shot learning because it alleviates the burden of ...manually collecting semantic attributes. Recent work has pursued this approach by exploring various ways of connecting the visual and text domains. In this paper, we revisit this idea by going further to consider one important factor: the textual representation is usually too noisy for the zero-shot learning application. This observation motivates us to design a simple yet effective zero-shot learning method that is capable of suppressing noise in the text. Specifically, we propose an l2,1-norm based objective function which can simultaneously suppress the noisy signal in the text and learn a function to match the text document and visual features. We also develop an optimization algorithm to efficiently solve the resulting problem. By conducting experiments on two large datasets, we demonstrate that the proposed method significantly outperforms those competing methods which rely on online information sources but with no explicit noise suppression. Furthermore, we make an in-depth analysis of the proposed method and provide insight as to what kind of information in documents is useful for zero-shot learning.
Rationale
Studies have shown the potential neuroprotective effect of xanthohumol, while whether xanthohumol has the ability of repairing cognitive impairment and its underlying mechanism still ...remains obscure.
Objectives
To unravel the mechanism of xanthohumol repairing cognitive impairment caused by estrogen deprivation.
Methods
C57BL/6 J female mice that underwent bilateral ovariectomy to establish cognitive decline model were randomly divided into three xanthohumol-treated groups and a saline-treated model group. For identifying the neuroprotective function of xanthohumol, Morris water maze (MWM) test and open field test (OFT) were conducted. After extracting total RNA of mouse hippocampus of different groups, mRNA-seq and microRNA (miRNA)-seq analysis were performed, and the differentially expressed miRNAs (DEMIs) and their target genes were further validated by qPCR. MiR-532-3p and its downstream gene
Mpped1
were screened as targets of xanthohumol. Influence of miR-532-3p/Mpped1 to cognitive ability was examined via MWM test and OFT after stereotactic brain injection of
Mpped1
overexpressed adeno-associated virus. The regulation of miR-532-3p on
Mpped1
was confirmed in hippocampal neuronal cell line HT22 by luciferase reporter gene assay.
Results
Xanthohumol treatment reversed the cognitive decline of OVX mice according to behavioral tests. By comparing miRNA levels of xanthohumol-treated groups with saline-treated group, we found that the main changed miRNAs were miR-122-5p, miR-532-3p, and miR-539-3p. Increased miR-532-3p in OVX mice was suppressed by xanthohumol treatment. Furthermore, the downstream gene of miR-532-3p,
Mpped1
, was also increased by xanthohumol and showed the capability of relieving cognitive impairment of OVX mice after overexpressed in hippocampus. The 3′ untranslated region of
Mpped1
was identified as the target region of miR-532-3p, and agomiR-532-3p remarkably reduced the expression of
Mpped1
mRNA.
Conclusions
Xanthohumol has the ability of repairing cognitive impairment through removing the inhibition of miR-532-3p on
Mpped1
in mouse hippocampus. This finding not only advances the understanding of neuroprotective mechanism of xanthohumol, but also provides novel treatment targets for dementia of postmenopausal women.
To address the huge labeling cost in large-scale point cloud semantic segmentation, we propose a novel hybrid contrastive regularization (HybridCR) framework in weakly-supervised setting, which ...obtains competitive performance compared to its fully-supervised counterpart. Specifically, HybridCR is the first framework to leverage both point consistency and employ contrastive regularization with pseudo labeling in an end-to-end manner. Fundamentally, HybridCR explicitly and effectively considers the semantic similarity between local neighboring points and global characteristics of 3D classes. We further design a dynamic point cloud augmentor to generate diversity and robust sample views, whose transformation parameter is jointly optimized with model training. Through extensive experiments, HybridCR achieves significant performance improvement against the SOTA methods on both indoor and outdoor datasets, e.g., S3DIS, ScanNet-V2, Semantic3D, and SemanticKITTI.
Liuwei Dihuang (LWDH) is a classic prescription that has been used to the treatment of “Kidney-Yin” deficiency syndrome for more than 1000 years in China. Recent studies have confirmed that LWDH can ...prevent the progression of renal fibrosis. Numerous studies have demonstrated the critical role that TRPC6 plays in the development of renal fibrosis. Due to the complex composition of LWDH and its remarkable therapeutic effect on renal fibrosis, it is possible to discover new active ingredients targeting TRPC6 for the treatment of renal fibrosis.
This study aimed to identify selective TRPC6 inhibitors from LWDH and evaluate their therapeutical effects on renal fibrosis.
Computer-aided drug design was used to screen the biologically active ingredients of LWDH, and their affinities to human TRPC6 protein were detected by microcalorimetry. TRPC6, TRPC3, and TRPC7 over-expressed HEK293 cells were constructed, and the selective activities of the compounds on TRPC6 were determined by measuring Ca2+i in these cells. To establish an in vitro model of renal fibrosis, human renal proximal tubular epithelial HK-2 cells were stimulated with TGF-β1. The therapeutic effects of LWDH compounds on renal fibrosis were then tested by detecting the related proteins. TRPC6 was knocked-down in HK-2 cells to investigate the effects of LWDH active ingredients on TRPC6. Finally, a unilateral ureteral obstruction model of renal fibrosis was established to test the therapeutic effect.
From hundreds of LWDH ingredients, 64 active components with oral bioavailability ≥30% and drug-likeness index ≥0.18 were acquired. A total of 10 active components were obtained by molecular docking with TRPC6 protein. Among them, 4 components had an affinity with TRPC6. Piperlonguminine (PLG) had the most potent affinity with TRPC6 and blocking effect on TRPC6-mediated Ca2+ entry. A 100 μM of PLG showed no detectable inhibition on TRPC1, TRPC3, TRPC4, TRPC5, or TRPC7-mediated Ca2+ influx into cells. In vitro results indicated that PLG concentration-dependently inhibited the abnormally high expression of α-smooth muscle actin (α-SMA), collagen I, vimentin, and TRPC6 in TGF-β1-induced HK-2 cells. Consistently, PLG also could not further inhibit TGF-β1-induced expressions of these protein biomarkers in TRPC6 knocked-down HK-2 cells. In vivo, PLG dose-dependently reduced urinary protein, serum creatinine, and blood urea nitrogen levels in renal fibrosis mice and markedly alleviated fibrosis and the expressions of α-SMA, collagen I, vimentin, and TRPC6 in kidney tissues.
Our results showed that PLG had anti-renal fibrosis effects by selectively inhibiting TRPC6. PLG might be a promising therapeutic agent for the treatment of renal fibrosis.
Display omitted
•PLG is an important anti-renal fibrosis active component in Liuwei Dihuang decoction.•PLG alleviates renal fibrosis in HK-2 cells and mice.•The therapeutic mechanism of PLG is through inhibiting TRPC6 and ultimately mitigating renal fibrosis.