3D point clouds reconstruction of mixed reality model based on medical imaging

This project intends to deploy image information such as magnetic resonance imaging(MRI)and digital subtraction angiography(DSA)com bined with stereo positioning system to scan,display and locate anatomical structures such as bones and joints in corpses and living bodies.We employ 3D point clouds as efficient representations of MRI,and propose point clouds denoising and inpainting leveraging on the field of graph signal processing.Virtual reality surgery for specific patients,such as arthroplasty,can be performed multiple times in the ward,and the calculated force is communicated to the operator through the force feedback device to obtain a feel that is close to the actual operation.

Recent advances of neural text generation:Core tasks,datasets,models and challenges

In recent years,deep neural network has achieved great success in solving many natural language processing tasks.Particularly,substantial progress has been made on neural text generation,which takes the linguistic and non-linguistic input,and generates natural language text.This survey aims to provide an up-to-date synthesis of core tasks in neural text generation and the architectures adopted to handle these tasks,and draw attention to the challenges in neural text generation.We first outline the mainstream neural text generation frameworks,and then introduce datasets,advanced models and challenges of four core text generation tasks in detail,including AMR-to-text generation,data-to-text generation,and two text-to-text generation tasks(i.e.,text summarization and paraphrase generation).Finally,we present future research directions for neural text generation.This survey can be used as a guide and reference for researchers and practitioners in this area.

Evolution Analysis of Societal Risk Events by Risk Maps

Event evolution analysis which provides an effective approach to capture the main context of a story from explosive increased news texts has become the critical basis for many real applications,such as crisis and emergency management and decision making.Especially,the development of societal risk events which may cause some possible harm to society or individuals has been heavily concerned by both the government and the public.In order to capture the evolution and trends of societal risk events,this paper presents an improved algorithm based on the method of information maps.It contains an event-level cluster generation algorithm and an evaluation algorithm.The main work includes:1)Word embedding representation is adopted and event-level clusters are chosen as nodes of the events evolution chains which may comprehensively present the underlying structure of events.Meanwhile,clusters that consist of risk-labeled events enable to illustrate how events evolve along the time with transitions of risks.2)One real-world case,the event of"Chinese Red Cross",is studied and a series of experiments are conducted.3)An evaluation algorithm is proposed on the basis of indicators of map construction without massive human-annotated dataset.Our approach for event evolution analysis automatically generates a visual evolution of societal risk events,displaying a clear and structural picture of events development.

Learning conditional photometric stereo with high-resolution features

Photometric stereo aims to reconstruct 3D geometry by recovering the dense surface orientation of a 3D object from multiple images under differing illumination.Traditional methods normally adopt simplified reflectance models to make the surface orientation computable.However,the real reflectances of surfaces greatly limit applicability of such methods to real-world objects.While deep neural networks have been employed to handle non-Lambertian surfaces,these methods are subject to blurring and errors,especially in high-frequency regions(such as crinkles and edges),caused by spectral bias:neural networks favor low-frequency representations so exhibit a bias towards smooth functions.In this paper,therefore,we propose a self-learning conditional network with multiscale features for photometric stereo,avoiding blurred reconstruction in such regions.Our explorations include:(i)a multi-scale feature fusion architecture,which keeps high-resolution representations and deep feature extraction,simultaneously,and(ii)an improved gradient-motivated conditionally parameterized convolution(GM-CondConv)in our photometric stereo network,with different combinations of convolution kernels for varying surfaces.Extensive experiments on public benchmark datasets show that our calibrated photometric stereo method outperforms the state-of-the-art.

Integrating Manifold Knowledge for Global Entity Linking with Heterogeneous Graphs

Entity Linking(EL)aims to automatically link the mentions in unstructured documents to corresponding entities in a knowledge base(KB),which has recently been dominated by global models.Although many global EL methods attempt to model the topical coherence among all linked entities,most of them failed in exploiting the correlations among manifold knowledge helpful for linking,such as the semantics of mentions and their candidates,the neighborhood information of candidate entities in KB and the fine-grained type information of entities.As we will show in the paper,interactions among these types of information are very useful for better characterizing the topic features of entities and more accurately estimating the topical coherence among all the referred entities within the same document.In this paper,we present a novel HEterogeneous Graph-based Entity Linker(HEGEL)for global entity linking,which builds an informative heterogeneous graph for every document to collect various linking clues.Then HEGEL utilizes a novel heterogeneous graph neural network(HGNN)to integrate the different types of manifold information and model the interactions among them.Experiments on the standard benchmark datasets demonstrate that HEGEL can well capture the global coherence and outperforms the prior state-of-the-art EL methods.