22:T539, Background: With significant advancement and demand for digital transformation, the digital twin has been gaining increasing attention as it is capable of establishing real-time mapping between physical space and virtual space. In this work, a shape-performance integrated digital twin solution is presented to predict the real-time biomechanics of the lumbar spine during human movement. Methods: A finite element model (FEM) of the lumbar spine was firstly developed using computed tomography (CT) and constrained by the body movement which was calculated by the inverse kinematics algorithm. The Gaussian process regression was utilized to train the predicted results and create the digital twin of the lumbar spine in real-time. Finally, a three-dimensional virtual reality system was developed using Unity3D to display and record the real-time biomechanics performance of the lumbar spine during body movement. Results: The evaluation results presented an agreement (R-squared > 0.8) between the real-time prediction from digital twin and offline FEM prediction. Conclusions: This approach provides an effective method of real-time planning and warning in spine rehabilitation. 23:T4b9,BERT-style models pre-trained on the general corpus (e.g., Wikipedia) and fine-tuned on specific task corpus, have recently emerged as breakthrough techniques in many NLP tasks: question answering, text classification, sequence labeling and so on. However, this tech- nique may not always work, especially for two scenarios: a corpus that contains very different text from the general corpus Wikipedia, or a task that learns embedding spacial distribution for a specific purpose (e.g., approximate nearest neighbor search). In this paper, to tackle the above two scenarios that we have encountered in an industrial e-commerce search system, we propose customized and novel pre-training tasks for two critical modules: user intent detec- tion and semantic embedding retrieval. The customized pre-trained models after fine-tuning, being less than 10% of BERT-base's size in order to be feasible for cost-efficient CPU serving, significantly improve the other baseline models: 1) no pre-training model and 2) fine-tuned model from the official pre-trained BERT using general corpus, on both offline datasets and online system. We have open sourced our datasets 1 for the sake of reproducibility and future works.24:T43b,One of the challenges of face recognition in surveillance is the low resolution of face region. Therefore many superresolution (SR) face reconstruction methods are proposed to produce a high-resolution face image from one or a set of low-resolution face images. However, existing dictionary learning based algorithms are sensitive to noise and very time-consuming. In this paper, we define and prove the multi-scale linear combination consistency. In order to improve the performance of SR, we propose a novel SR face reconstruction method based on nonlocal similarity and multi-scale linear combination consistency (NLS-MLC). We further proposed a new recognition approach for very low resolution face images based on resolution scale invariant feature (RSIF). A series of experiments are conducted on two public face image databases to test feasibility of our proposed methods. Experimental results show that the proposed SR method is more robust and computationally effective in face hallucination, and the recognition accuracy of RSIF is higher than some state-of-art algorithms.

#	Concept	H-Index	Publications	Citations
1	Engineering	5	5	134
2	Education	1	1	61
3	Language Studies	1	1	17
4	Computer Engineering	1	1	61
5	Data Science	1	1	61

	Year	Citations

About

Top Concepts