Search

Abstract Shape descriptors have recently gained popularity in shape matching, statistical shape modeling, etc. Their discriminative ability and efficiency play a decisive role in these tasks. In this paper, we first propose a novel handcrafted anisotropic spectral descriptor using Chebyshev polynomials, called the anisotropic Chebyshev descriptor (ACD); it can effectively capture shape features in multiple directions. The ACD inherits many good characteristics of spectral descriptors, such as being intrinsic, robust to changes in surface discretization, etc. Furthermore, due to the orthogonality of Chebyshev polynomials, the ACD is compact and can disambiguate intrinsic symmetry since several directions are considered. To improve the ACD’s discrimination ability, we construct a Chebyshev spectral manifold convolutional neural network (CSMCNN) that optimizes the ACD and produces a learned ACD. Our experimental results show that the ACD outperforms existing state-of-the-art handcrafted descriptors. The combination of the ACD and the CSMCNN is better than other state-of-the-art learned descriptors in terms of discrimination, efficiency, and robustness to changes in shape resolution and discretization.

Abstract Existing deep unfolding methods unroll an optimization algorithm with a fixed number of steps, and utilize convolutional neural networks (CNNs) to learn data-driven priors. However, their performance is limited for two main reasons. Firstly, priors learned in deep feature space need to be converted to the image space at each iteration step, which limits the depth of CNNs and prevents CNNs from exploiting contextual information. Secondly, existing methods only learn deep priors at the single full-resolution scale, so ignore the benefits of multi-scale context in dealing with high level noise. To address these issues, we explicitly consider the image denoising process in the deep feature space and propose the deep unfolding multi-scale regularizer network (DUMRN) for image denoising. The core of DUMRN is the feature-based denoising module (FDM) that directly removes noise in the deep feature space. In each FDM, we construct a multi-scale regularizer block to learn deep prior information from multi-resolution features. We build the DUMRN by stacking a sequence of FDMs and train it in an end-to-end manner. Experimental results on synthetic and real-world benchmarks demonstrate that DUMRN performs favorably compared to state-of-the-art methods.

In traditional deep functional maps for non-rigid shape correspondence, estimating a functional map including high-frequency information requires enough linearly independent features via the least square method, which is prone to be violated in practice, especially at an early stage of training, or costly post-processing, e.g. ZoomOut. In this paper, we propose a novel method called RFMNet (Robust Deep Functional Map Networks), which jointly considers training stability and more geometric shape features than previous works. We directly first produce a pointwise map by resorting to optimal transport and then convert it to an initial functional map. Such a mechanism mitigates the requirements for the descriptor and avoids the training instabilities resulting from the least square solver. Benefitting from the novel strategy, we successfully integrate a state-of-the-art geometric regularization for further optimizing the functional map, which substantially filters the initial functional map. We show our novel computing functional map module brings more stable training even under encoding the functional map with high-frequency information and faster convergence speed. Considering the pointwise and functional maps, an unsupervised loss is presented for penalizing the correspondence distortion of Delta functions between shapes. To catch discretization-resistant and orientation-aware shape features with our network, we utilize DiffusionNet as a feature extractor. Experimental results demonstrate our apparent superiority in correspondence quality and generalization across various shape discretizations and different datasets compared to the state-of-the-art learning methods.

Abstract Specular highlight detection and removal is a fundamental problem in computer vision and image processing. In this paper, we present an efficient end-to-end deep learning model for automatically detecting and removing specular highlights in a single image. In particular, an encoder—decoder network is utilized to detect specular highlights, and then a novel Unet-Transformer network performs highlight removal; we append transformer modules instead of feature maps in the Unet architecture. We also introduce a highlight detection module as a mask to guide the removal task. Thus, these two networks can be jointly trained in an effective manner. Thanks to the hierarchical and global properties of the transformer mechanism, our framework is able to establish relationships between continuous self-attention layers, making it possible to directly model the mapping between the diffuse area and the specular highlight area, and reduce indeterminacy within areas containing strong specular highlight reflection. Experiments on public benchmark and real-world images demonstrate that our approach outperforms state-of-the-art methods for both highlight detection and removal tasks.

Severe cerebral ischemia/reperfusion injury has been shown to induce high-level autophagy and neuronal death. Therefore, it is extremely important to search for a target that inhibits autophagy activation. Long non-coding RNA MEG3 participates in autophagy. However, it remains unclear whether it can be targeted to regulate cerebral ischemia/reperfusion injury. Our results revealed that in oxygen and glucose deprivation/reoxygenation-treated HT22 cells, MEG3 expression was obviously upregulated, and autophagy was increased, while knockdown of MEG3 expression greatly reduced autophagy. Furthermore, MEG3 bound miR-181c-5p and inhibited its expression, while miR-181c-5p bound to autophagy-related gene ATG7 and inhibited its expression. Further experiments revealed that mir-181c-5p overexpression reversed the effect of MEG3 on autophagy and ATG7 expression in HT22 cells subjected to oxygen and glucose deprivation/reoxygenation. In vivo experiments revealed that MEG3 knockdown suppressed autophagy, infarct volume and behavioral deficits in cerebral ischemia/reperfusion mice. These findings suggest that MEG3 knockdown inhibited autophagy and alleviated cerebral ischemia/reperfusion injury through the miR-181c-5p/ATG7 signaling pathway. Therefore, MEG3 can be considered as an intervention target for the treatment of cerebral ischemia/reperfusion injury. This study was approved by the Animal Ethics Committee of the First Affiliated Hospital of Zhengzhou University, China (approval No. XF20190538) on January 4, 2019.

Abstract Recent learning-based approaches show promising performance improvement for the scene text removal task but usually leave several remnants of text and provide visually unpleasant results. In this work, a novel end-to-end framework is proposed based on accurate text stroke detection. Specifically, the text removal problem is decoupled into text stroke detection and stroke removal; we design separate networks to solve these two subproblems, the latter being a generative network. These two networks are combined as a processing unit, which is cascaded to obtain our final model for text removal. Experimental results demonstrate that the proposed method substantially outperforms the state-of-the-art for locating and erasing scene text. A new large-scale real-world dataset with 12,120 images has been constructed and is being made available to facilitate research, as current publicly available datasets are mainly synthetic so cannot properly measure the performance of different methods.

Cholestasis is a clinical syndrome triggered by the accumulation and aggregation of bile acids by subsequent inflammatory responses. The present study investigated the protective effect of glycyrrhetinic acid (GA) on the cholestatic liver injury induced by lithocholic acid (LCA) from both anti-inflammatory and choleretic mechanistic standpoints. Male C57BL/6 mice were treated with LCA twice daily for 4 days to induce intrahepatic cholestasis. GA (50 mg/kg) and pregnenolone 16α-carbonitrile (PCN, 45 mg/kg) were intraperitoneally injected 3 days before and throughout the administration of LCA, respectively. Plasma biochemical indexes were determined by assay kits, and hepatic bile acids were quantified by LC-MS/MS. Hematoxylin and eosin staining of liver sections was performed for pathological examination. Protein expression of the TLRs/NF-κB pathway and the mRNA levels of inflammatory cytokines and chemokines were examined by Western blotting and PCR, respectively. Finally, the hepatic expression of pregnane X receptor (PXR) and farnesoid X receptor (FXR) and their target genes encoding metabolic enzymes and transporters was evaluated. GA significantly reversed liver necrosis and decreased plasma ALT and ALP activity. Plasma total bile acids, total bilirubin, and hepatic bile acids were also remarkably preserved. More importantly, the recruitment of inflammatory cells to hepatic sinusoids was alleviated. Additionally, the protein expression of TLR2, TLR4, and p-NF-κBp65 and the mRNA expression of CCL2, CXCL2, IL-1β, IL-6, and TNF-α were significantly decreased. Moreover, GA significantly increased the expression of hepatic FXR and its target genes, including BSEP, MRP3, and MRP4. In conclusion, GA protects against LCA-induced cholestatic liver injury by inhibiting the TLR2/NF-κB pathway and upregulating hepatic FXR expression.

Abstract A discriminative local shape descriptor plays an important role in various applications. In this paper, we present a novel deep learning framework that derives discriminative local descriptors for deformable 3D shapes. We use local “geometry images” to encode the multi-scale local features of a point, via an intrinsic parameterization method based on geodesic polar coordinates. This new parameterization provides robust geometry images even for badly-shaped triangular meshes. Then a triplet network with shared architecture and parameters is used to perform deep metric learning; its aim is to distinguish between similar and dissimilar pairs of points. Additionally, a newly designed triplet loss function is minimized for improved, accurate training of the triplet network. To solve the dense correspondence problem, an efficient sampling approach is utilized to achieve a good compromise between training performance and descriptor quality. During testing, given a geometry image of a point of interest, our network outputs a discriminative local descriptor for it. Extensive testing of non-rigid dense shape matching on a variety of benchmarks demonstrates the superiority of the proposed descriptors over the state-of-the-art alternatives.

Tumor migration and invasion are key pathological processes that contribute to cell metastasis as well as treatment failure in patients with malignant tumors. However, the mechanisms governing tumor cell migration remain poorly understood. By analyzing the tumor-related database and tumor cell lines, we found that preoptic regulatory factor-2 (Porf-2) is downexpressed in both neuroblastoma and glioma. Using in vitro assays, our data demonstrated that the expression of Porf-2 inhibits tumor cell migration both in neuroblastoma and glioma cell lines. Domain-mutated Porf-2 plasmids were then constructed, and it was found that the GAP domain, which plays a role in the inactivation of Rac1, is the functional domain for inhibiting tumor cell migration. Furthermore, by screening potential downstream effectors, we found that Porf-2 can reduce MMP-2 and MMP-9 expression. Overexpression of MMP-2 blocked the inhibitory effect of Porf-2 in tumor cell migration both in vitro and in vivo. Taken together, we show for the first time that Porf-2 is capable of suppressing tumor cell migration via its GAP domain and the downregulation of MMP-2/9, suggesting that targeting Porf-2 could be a promising therapeutic strategy for nervous system tumors.

Abstract With the development of biomolecular modeling and simulation, especially implicit solvent modeling, higher requirements are set for the stability, efficiency and mesh quality of molecular mesh generation software. In this review, we summarize the recent works in biomolecular mesh generation and molecular visualization. First, we introduce various definitions of molecular surface and corresponding meshing software. Second, as the mesh quality significantly influences biomolecular simulation, we investigate some remeshing methods in the fields of computer graphics and molecular modeling. Then, we show the application of biomolecular mesh in the boundary element method (BEM) and the finite element method (FEM). Finally, to conveniently visualize the numerical results based on the mesh, we present two types of molecular visualization systems.

Abstract Surface remeshing is widely required in modeling, animation, simulation, and many other computer graphics applications. Improving the elements’ quality is a challenging task in surface remeshing. Existing methods often fail to efficiently remove poor-quality elements especially in regions with sharp features. In this paper, we propose and use a robust segmentation method followed by remeshing the segmented mesh. Mesh segmentation is initiated using an existing Live-wire interaction approach and is further refined using local mesh operations. The refined segmented mesh is finally sent to the remeshing pipeline, in which each mesh segment is remeshed independently. An experimental study compares our mesh segmentation method as well as remeshing results with representative existing methods. We demonstrate that the proposed segmentation method is robust and suitable for remeshing.

Fiducial markers are images or landmarks placed in real environment, typically used for pose estimation and camera tracking. Reliable fiducials are strongly desired for many augmented reality (AR) applications, but currently there is no systematic method to design highly reliable fiducials. In this paper, we present fiducial marker optimizer (FMO), a tool to optimize the design attributes of ARToolKit markers, including black to white (B:W) ratio, edge sharpness, and information complexity, and to reduce inter-marker confusion. For these operations, the FMO provides a user friendly interface at the front-end and specialized image processing algorithms at the back-end. We tested manually designed markers and FMO optimized markers in ARToolKit and found that the latter were more robust. The FMO will be used for designing highly reliable fiducials in easy to use fashion. It will improve the application's performance, where it is used.