面向流场可视化的沉浸式虚拟现实协作框架郑志文 1,2) , 杨超 2,3) , 韩永国 1) , 杨洋 4)* , 吴亚东 5) , 王昉 2,3) , 张晓蓉 1) 1) (西南科技大学计算机科学与技术学院 绵阳 621010) 2) (中国空气动力研究与发展中心计算空气动力研究所 绵阳 621000) 3) (空气动力学国家重点实验室 绵阳 621000) 4) (四川省计算机研究院 成都 610041) 5) (四川轻化工大学计算机科学与工程学院 自贡 643002)

No abstract

No abstract

Transportation system has time-varying, coupling and nonlinear dynamic characteristics. Traffic flow forecast is one of the key technologies of traffic guidance. It is very difficult to accurately forecast them effectively. This paper has analyzed the complexity and the evaluation index of urban transportation network and has proposed the forecasting model of the hybrid algorithm based on chaos immune knowledge. First of all, the chaos knowledge is introduced into the topology structure of immune network, so as to obtain the matching predictive values and knowledge base. Secondly, this algorithm can dynamically control and adjusted the regional search speed and can fuse the information obtained by the chaos and immune algorithm, in order to realize the short-term traffic flow forecast. Finally, the simulation experiment shows that the traffic flow forecasting error obtained by the method is small, feasible and effective and can better meet the needs of the traffic guidance system.

Serious feature heterogeneity and semantic gaps exist between multi-source heterogeneous data, and existing cross-modal retrieval methods cannot effectively extract common semantic and complementary information between multi-source heterogeneous data. In this regard, an adversarial cross-modal retrieval method that fuses collaborative attention networks is proposed. The method addresses two major challenges in the cross-modal retrieval process, firstly, an information extraction algorithm based on the cooperative attention mechanism is designed, and secondly, the cooperative attention network is combined with the adversarial subspace learning algorithm to enhance the complementary capability of information in the feature subspace. The experimental results show that the proposed method has better retrieval results than similar cross-modal retrieval methods in terms of MAP metrics.

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ˈ EXW PD[ J [ LV XQNQRZQ 7KH PD[LPXP J [ LV UHSODFHG E\ WKH FXUUHQW JHQHUDWLRQ7KH

No abstract

Recently, digital image inpainting technique based on variational and partial differential equation has been researched widely. In this paper, image inpainting is applied to image encoding and a novel scheme of encoding is proposed. In the encoding, applying an edge detector to detect the edge. In addition, the four boundaries of the image also belong to the edge collection; next, generating small neighborhoods of the edge collection to get an edge-extended image, then encoding the addresses of the edge-extended image and using high bit rate to accurately code the gray values of it. So, a high compression rate is achieved. Unlike the traditional decoding, here the decoding is realized by a harmonic inpainting model to achieve a good quality of reconstructed image. At the end, a lot of images are simulated. These simulated results show that with this scheme proposed, one can get a good quality of reconstructed image in a high compression rate. Keywords-image encodingt;image decoding; image inpaintingI.