DHM-Net: Deep Hypergraph Modeling for Robust Feature Matching.

We present a novel deep hypergraph modeling architecture (called DHM-Net) for feature matching in this paper. Our network focuses on learning reliable correspondences between two sets of initial feature points by establishing a dynamic hypergraph structure that models group-wise relationships and assigns weights to each node. Compared to existing feature matching methods that only consider pair-wise relationships via a simple graph, our dynamic hypergraph is capable of modeling nonlinear higher-order group-wise relationships among correspondences in an interaction capturing and attention representation learning fashion.

Highly Enhanced Mechanical, Thermal, and Crystallization Performance of PLA/PBS Composite by Glass Fiber Coupling Agent Modification.

To improve the toughness and heat resistance of polylactic acid (PLA), polybutylene succinate (PBS) was sufficiently blended with PLA as the base matrix, and the glass fiber (GF) that was modified with 3-aminopropyltriethoxysilane (KF-GF) was added as the reinforcement. The results demonstrated a noteworthy boost in both mechanical and heat resistance properties when employing KH-GF, in comparison to pristine GF. When the content of KH-GF reached 20%, the tensile, flexural, and IZOD impact strength of the composites were 65.