Virtual Reality Aided High-Quality 3D Reconstruction by Remote Drones

Artificial intelligence including deep learning and 3D reconstruction methods is changing the daily life of people. Now, an unmanned aerial vehicle that can move freely in the air and avoid harsh ground conditions has been commonly adopted as a suitable tool for 3D reconstruction. The traditional 3D reconstruction mission based on drones usually consists of two steps: image collection and offline post-processing. But there are two problems: one is the uncertainty of whether all parts of the target object are covered, and another is the tedious post-processing time. Inspired by modern deep learning methods, we build a telexistence drone system with an onboard deep learning computation module and a wireless data transmission module that perform incremental real-time dense reconstruction of urban cities by itself. Two technical contributions are proposed to solve the preceding issues. First, based on the popular depth fusion surface reconstruction framework, we combine it with a visual-inertial odometry estimator that integrates the inertial measurement unit and allows for robust camera tracking as well as high-accuracy online 3D scan. Second, the capability of real-time 3D reconstruction enables a new rendering technique that can visualize the reconstructed geometry of the target as navigation guidance in the HMD. Therefore, it turns the traditional path-planning-based modeling process into an interactive one, leading to a higher level of scan completeness. The experiments in the simulation system and our real prototype demonstrate an improved quality of the 3D model using our artificial intelligence leveraged drone system.

Effective structural unit analysis in hexagonal close-packed alloys – reconstruction of parent β microstructures and crystal orientation post-processing analysis

Materials with an allotropic phase transformation can form microstructures where grains have orientation relationships determined by the transformation history. These microstructures influence the final material properties. In zirconium alloys, there is a solid-state body-centred cubic (b.c.c.) to hexagonal close-packed (h.c.p.) phase transformation, where the crystal orientations of the h.c.p. phase can be related to the parent b.c.c. structure via the Burgers orientation relationship (BOR). In the present work, a reconstruction code, developed for steels and which uses a Markov chain clustering algorithm to analyse electron backscatter diffraction maps, is adapted and applied to the h.c.p./b.c.c. BOR. This algorithm is released as open-source code (via github, as ParentBOR). The algorithm enables new post-processing of the original and reconstructed data sets to analyse the variants of the h.c.p. α phase that are present and understand shared crystal planes and shared lattice directions within each parent β grain; it is anticipated that this will assist in understanding the transformation-related deformation properties of the final microstructure. Finally, the ParentBOR code is compared with recently released reconstruction codes implemented in MTEX to reveal differences and similarities in how the microstructure is described.

Neural Arabic Text Diacritization: State-of-the-Art Results and a Novel Approach for Arabic NLP Downstream Tasks

In this work, we present several deep learning models for the automatic diacritization of Arabic text. Our models are built using two main approaches, viz. Feed-Forward Neural Network (FFNN) and Recurrent Neural Network (RNN), with several enhancements such as 100-hot encoding, embeddings, Conditional Random Field (CRF), and Block-Normalized Gradient (BNG). The models are tested on the only freely available benchmark dataset and the results show that our models are either better or on par with other models even those requiring human-crafted language-dependent post-processing steps, unlike ours. Moreover, we show how diacritics in Arabic can be used to enhance the models of downstream NLP tasks such as Machine Translation (MT) and Sentiment Analysis (SA) by proposing novel Translation over Diacritization (ToD) and Sentiment over Diacritization (SoD) approaches.

Multi-interest Diversification for End-to-end Sequential Recommendation

Sequential recommenders capture dynamic aspects of users’ interests by modeling sequential behavior. Previous studies on sequential recommendations mostly aim to identify users’ main recent interests to optimize the recommendation accuracy; they often neglect the fact that users display multiple interests over extended periods of time, which could be used to improve the diversity of lists of recommended items. Existing work related to diversified recommendation typically assumes that users’ preferences are static and depend on post-processing the candidate list of recommended items. However, those conditions are not suitable when applied to sequential recommendations. We tackle sequential recommendation as a list generation process and propose a unified approach to take accuracy as well as diversity into consideration, called multi-interest, diversified, sequential recommendation . Particularly, an implicit interest mining module is first used to mine users’ multiple interests, which are reflected in users’ sequential behavior. Then an interest-aware, diversity promoting decoder is designed to produce recommendations that cover those interests. For training, we introduce an interest-aware, diversity promoting loss function that can supervise the model to learn to recommend accurate as well as diversified items. We conduct comprehensive experiments on four public datasets and the results show that our proposal outperforms state-of-the-art methods regarding diversity while producing comparable or better accuracy for sequential recommendation.

A Review on Manufacturing and Post-Processing Technology of Vascular Stents

Percutaneous coronary intervention (PCI) with stent implantation is one of the most effective treatments for cardiovascular diseases (CVDs). However, there are still many complications after stent implantation. As a medical device with a complex structure and small size, the manufacture and post-processing technology greatly impact the mechanical and medical performances of stents. In this paper, the development history, material, manufacturing method, and post-processing technology of vascular stents are introduced. In particular, this paper focuses on the existing manufacturing technology and post-processing technology of vascular stents and the impact of these technologies on stent performance is described and discussed. Moreover, the future development of vascular stent manufacturing technology will be prospected and proposed.

Bathymetric Survey of the St. Anthony Channel (Croatia) Using Multibeam Echosounders (MBES)—A New Methodological Semi-Automatic Approach of Point Cloud Post-Processing

Multibeam echosounders (MBES) have become a valuable tool for underwater floor mapping. However, MBES data are often loaded with different measurement errors. This study presents a new user-friendly and methodological semi-automatic approach of point cloud post-processing error removal. The St. Anthony Channel (Croatia) was selected as the research area because it is regarded as one of the most demanding sea or river passages in the world and it is protected as a significant landscape by the Šibenik-Knin County. The two main objectives of this study, conducted within the Interreg Italy–Croatia PEPSEA project, were to: (a) propose a methodological framework that would enable the easier and user-friendly identification and removal of the errors in MBES data; (b) create a high-resolution integral model (MBES and UAV data) of the St. Anthony Channel for maritime safety and tourism promotion purposes. A hydrographic survey of the channel was carried out using WASSP S3 MBES while UAV photogrammetry was performed using Matrice 210 RTK V2. The proposed semi-automatic post-processing of the MBES acquired point cloud was completed in the Open Source CloudCompare software following five steps in which various point filtering methods were used. The reduction percentage in points after the denoising process was 14.11%. Our results provided: (a) a new user-friendly methodological framework for MBES point filtering; (b) a detailed bathymetric map of the St. Anthony Channel with a spatial resolution of 50 cm; and (c) the first integral (MBES and UAV) high-resolution model of the St. Anthony Channel. The generated models can primarily be used for maritime safety and tourism promotion purposes. In future research, ground-truthing methods (e.g., ROVs) will be used to validate the generated models.