Learning-Based Intrinsic Reflectional Symmetry Detection.

Reflectional symmetry is a ubiquitous pattern in nature. Previous works usually solve this problem by voting or sampling, suffering from high computational cost and randomness. In this article, we propose a learning-based approach to intrinsic reflectional symmetry detection. Instead of directly finding symmetric point pairs, we parametrize this self-isometry using a functional map matrix, which can be easily computed given the signs of Laplacian eigenfunctions under the symmetric mapping. Therefore, we manually label the eigenfunction signs for a variety of shapes and train a novel neural network to predict the sign of each eigenfunction under symmetry. Our network aims at learning the global property of functions and consequently converts the problem defined on the manifold to the functional domain. By disentangling the prediction of the matrix into separated bases, our method generalizes well to new shapes and is invariant under perturbation of eigenfunctions. Through extensive experiments, we demonstrate the robustness of our method in challenging cases, including different topology and incomplete shapes with holes. By avoiding random sampling, our learning-based algorithm is over 20 times faster than state-of-the-art methods, and meanwhile, is more robust, achieving higher correspondence accuracy in commonly used metrics.

Synthesizing Mesh Deformation Sequences With Bidirectional LSTM.

Synthesizing realistic 3D mesh deformation sequences is a challenging but important task in computer animation. To achieve this, researchers have long been focusing on shape analysis to develop new interpolation and extrapolation techniques. However, such techniques have limited learning capabilities and therefore often produce unrealistic deformation. Although there are already networks defined on individual meshes, deep architectures that operate directly on mesh sequences with temporal information remain unexplored due to the following major barriers: irregular mesh connectivity, rich temporal information, and varied deformation. To address these issues, we utilize convolutional neural networks defined on triangular meshes along with a shape deformation representation to extract useful features, followed by long short-term memory (LSTM) that iteratively processes the features. To fully respect the bidirectional nature of actions, we propose a new share-weight bidirectional scheme to better synthesize deformations. An extensive evaluation shows that our approach outperforms existing methods in sequence generation, both qualitatively and quantitatively.

Learning on 3D Meshes With Laplacian Encoding and Pooling.

3D models are commonly used in computer vision and graphics. With the wider availability of mesh data, an efficient and intrinsic deep learning approach to processing 3D meshes is in great need. Unlike images, 3D meshes have irregular connectivity, requiring careful design to capture relations in the data. To utilize the topology information while staying robust under different triangulations, we propose to encode mesh connectivity using Laplacian spectral analysis, along with mesh feature aggregation blocks (MFABs) that can split the surface domain into local pooling patches and aggregate global information amongst them. We build a mesh hierarchy from fine to coarse using Laplacian spectral clustering, which is flexible under isometric transformations. Inside the MFABs there are pooling layers to collect local information and multi-layer perceptrons to compute vertex features of increasing complexity. To obtain the relationships among different clusters, we introduce a Correlation Net to compute a correlation matrix, which can aggregate the features globally by matrix multiplication with cluster features. Our network architecture is flexible enough to be used on meshes with different numbers of vertices. We conduct several experiments including shape segmentation and classification, and our method outperforms state-of-the-art algorithms for these tasks on the ShapeNet and COSEG datasets.

Promoter hypermethylation-mediated downregulation of miR-770 and its host gene MEG3, a long non-coding RNA, in the development of gastric cardia adenocarcinoma.

Maternally expressed gene 3 (MEG3) is an imprinted gene located at 14q32 which encodes an lncRNA and is downregulated in an expanding list of cancer cell lines and primary human cancers. The miR-770 is transcribed from the intronic sequence of MEG3 and MEG3 may be the host gene for miR-770. However, the biological role of MEG3 and miR-770 in gastric cardia adenocarcinoma (GCA) development and prognosis is poorly defined. The present study was to investigate the function and methylation status of MEG3 in GCA, and further to detect the functional association of miR-770 and its host gene MEG3 in GCA carcinogenesis and prognosis. MEG3 and miR-770 was significantly downregulated in GCA patients and cell lines, and their expression was associated with TNM stage and lymph node metastasis. Overexpression of MEG3 and miR-770 inhibited gastric cancer cell proliferation and invasion in vitro. Furthermore, the expression level of MEG3 and miR-770 was significantly increased in cancer cells after treated with 5-Aza-dC. The aberrant hypermethylation of proximal promoter and enhancer region of MEG3 was detected in GCA tissues. In addition, the proximal promoter and enhancer region hypermethylation and dysregulation of MEG3 and miR-770 were associated with poorer GCA patients' survival. These findings suggest that miR-770 and its host gene MEG3 may play tumor suppressor role and hypermethylation of proximal promoter and enhancer region may be one of the critical mechanisms in inactivation of MEG3 and miR-770 in GCA development. MEG3 and miR-770 may be used as potential biomarkers in predicting GCA patients' prognosis.