In real training, the training conditions are often undesirable, and the use of equipment is severely limited. These problems can be solved by virtual practical training, which breaks the limit of space, lowers the training cost, while ensuring the training quality. However, the existing methods work poorly in image reconstruction, because they fail to consider the fact that the environmental perception of actual scene is strongly regular by nature. Therefore, this paper investigates the three-dimensional (3D) image reconstruction for virtual talent training scene. Specifically, a fusion network model was deigned, and the deep-seated correlation between target detection and semantic segmentation was discussed for images shot in two-dimensional (2D) scenes, in order to enhance the extraction effect of image features. Next, the vertical and horizontal parallaxes of the scene were solved, and the depth-based virtual talent training scene was reconstructed three dimensionally, based on the continuity of scene depth. Finally, the proposed algorithm was proved effective through experiments.

With the development of information technology, how to scientifically and properly organizing and guiding learners to learn actively and efficiently has become a research subject for domestic and foreign scholars. However, existing research on online learning behaviours studied little about learning attitudes, learning preferences, student-student interaction, teacher-student interaction and so on. To this end, this paper studies the features of group online learning behaviours based on data mining. In this paper, a K-means-based group online learning behaviour feature selection model and an AdaBoost-based group online learning behaviour classification model were constructed, and the processing methods, execution processes and algorithm functions of the two models were described in detail. Finally, the effectiveness of the constructed models was verified through an experiment.