Neural Face Swapping technique constitutes an image processing approach extensively adopted in social media platforms for short-form video enterprises to generate visually compelling and intriguing facial replacement videos. The technology appeals to the depth of the user’s experience and has promising applications. Nevertheless, the complexity in extracting facial details against various backgrounds poses a significant challenge to facial reenactment. The core manifests itself above controlling facial details, a challenge that constrains leaps in neural face swapping technology. To address the limitations of this methodology, we have enhanced the prevailing facial reenactment model to provide users with more precise synthetic videos. We will detail our enhancement procedure by strengthening the network; and optimizing the loss function to improve the quality of the model outputs, the model apprehends subtler facial details and accomplishes high quality synthesis. Finally, we compare with the more dominant synthesis algorithms of the day, the reinforced network has shown significant improvements, both visually and in terms of data for different evaluation metrics.