Research on traditional oil painting innovation and creation technology assisted by artificial intelligence algorithm

With the popularization and depth of the development of artificial intelligence in all walks of life, the field of art creation is also inspired by artificial intelligence. In this paper, artificial intelligence algorithms are embedded in the creation of oil paintings to assist the new era creation of oil paintings through the innovation of traditional creation techniques. The oil painting is mathematically modeled, and a model-based deep deterministic gradient strategy algorithm (DDPG) is proposed on the basis of reinforcement learning algorithms to simulate multi-brushstroke painting for oil painting. The oil painting creation assistance algorithm (DDPG) in this paper is compared with other oil painting assistance generation algorithms to explore the rendering performance and creation effect of the DDPG algorithm in this paper. The accuracy of the generated strokes of this paper’s DDPG renderer is 83.38%, which exceeds all the compared models and obtains the best rendering effect. The DDPG model in this paper has a total of 643 high-score ballots, with an average score of 808.2. It outperforms other models in the subjective evaluation of human visual perception.The DDPG method achieves the optimal results in assisting the generation of different styles of oil paintings, and the values of SSIM, MSE, PSNR, and LPIPS are 0.60, 0.0182, 12.80, and 0.3228, respectively, which are significantly better than the other oil painting assisted generation methods.