Neural Style Transfer For Artistic Photo-to-Painting Translation

Deep learning was used to create a neural style transfer framework for artistic phototopainting translation. The system balances content retention and stylistic change for real-time performance.  Next steps include integrating diffusion-based stylisation models, allowing region-based selective stylisation, adding user-controlled style intensity parameters, and optimising models for resource-constrained devices.

[1]   Y. Chen et al., “TransformerBased Neural Style Transfer for Artistic Image Synthesis,” IEEE Transactions on Pattern Analysis and Machine Intelligence, 2024. [2]   M. Kumar et al., “AttentionGuided Artistic Style Transfer Using Deep Networks,” IEEE Transactions on Artificial Intelligence, 2025.  [3]      L. Zhang et al., “Adaptive Instance Normalization for Efficient Neural Style Transfer,” IEEE Access, 2024. [4]      R. Patel et al., “Arbitrary MultiStyle Transfer with Dynamic Feature Modulation,” IEEE Transactions on Image Processing, 2025.  [5]     H. Lee et al., “Adversarial Perceptual Learning for Realistic Artistic Image Translation,” IEEE Access, 2026.  [6]   A. Singh et al., “Lightweight Neural Style Transfer Networks for Mobile Applications,” IEEE Internet of Things Journal, 2025. [7]        T. Nguyen et al., “CrossAttention Mechanisms for Visual Style Manipulation,” IEEE Transactions on Artificial Intelligence, 2023.  [8]  P. Wang et al., “HighFidelity Artistic Image Generation Using Deep Learning,” IEEE Access, 2024.  [9]  K. Roy et al., “Perceptual Loss Optimization in Visual Style Transfer,” IEEE Transactions on Neural Networks and Learning Systems, 2025. [10]   J. Brown et al., “Creative AI Systems for Artistic Image Synthesis,” IEEE Access, 2026.