Abstract
Virtual reality is a computer technology that produces a simulated environment. It is completely immersive and gives users the viewpoint that they are somewhere else. In recent times, it has become a highly interactive and visualization tool that has gained interest among educators and scholars. Art learning is a teaching-learning approach that is dependent on learning "through the arts" and "with the arts;" it can be a procedure in which art develops the medium of teaching-learning and an important model in some subjects of the curriculum. In this work, we develop a grey wolf optimization with the residual network form of virtual reality application for environmental art learning (GWORN-EAL) technique. It aims to provide metacognitive actions to improve environmental art learning for young children or adults. The GWORN-EAL technique is mainly based on the stimulation of particular features of the target painting over a default image. The color palette of the recognized image of the Fauve painter was mapped to the target image using the Fauve vision of the painter and represented by vivid colors. For optimal hyperparameter tuning of the ResNet model, the GWO algorithm is employed. The experimental results indicated that the GWORN-EAL technique has accomplished effectual outcomes in several aspects. A brief experimental study highlighted the improvement of the GWORN-EAL technique compared to existing models.