The dynamic field of fashion design increasingly demands innovative approaches to streamline style creation and enhance designer creativity. Traditional methods relying on manual sketching and coloring are time-consuming and limit rapid exploration of styles. Recognizing the challenges in automatic fashion style generation, particularly the lack of high-fidelity image recognition and realistic style synthesis, the research proposes a novel deep learning (DL) framework named the automated fashion style generation model based on image recognition techniques. To address the identified research gap, the research introduces a novel Vision Transformer-driven Style Generative Adversarial Network (ViT-StyleGAN), aiming to transform raw fashion imagery into diverse, high-quality styles. For this research, a comprehensive dataset comprising fashion items was assembled to ensure rich stylistic diversity. Data preprocessing involved noise reduction and clothing region segmentation using a Mask R-CNN variant, ensuring clean inputs for the model. Feature extraction was performed using an enhanced Vision Transformer (ViT) encoder to capture detailed spatial representations. The proposed framework processes the gathered data by first extracting robust features; it is subsequently used to create innovative fashion designs using the StyleGAN-based generator. ViT-StyleGAN effectively leverages global visual context and fine-grained texture generation, optimizing outputs for realism and creativity. The ViTStyleGAN was evaluated with the YOLOv5s models that were trained for 100 and 300 epochs. A mAP (98.8%), precision (98.5%), recall (98%), F1-score (98.3%), and FPS of 42 were attained in 100 epochs. With a mAP (99.3%), a precision (98.2%), a recall (98.8 %), an F1-score (98.5%), and FPS of 42 attained in 300 epochs, the ViT-StyleGAN fared better than both in terms of style realism, recognition accuracy, and speed. This framework establishes a new direction in automated fashion style generation, providing a powerful tool for designers to accelerate creativity while preserving personal esthetic preferences.