With the development of e-commerce, the number of consumers engaging in online clothing shopping has rapidly increased. To attract more attention to their fashion apparel, retailers are required to provide high-quality clothing display images that showcase rich details. Traditional virtual try-on techniques often lead to a loss of clothing details due to unnatural deformations of the garments. To address this issue, we propose a framework called Mannequin2Real+ for transforming mannequin images into photorealistic model images while preserving complete clothing details. The designed framework consists of two stages: a coarse-grained human body generation module and a refined-clothing human body alignment module. The model first generates a coarse-grained photorealistic model based on the mannequin’s pose and the target model’s facial features while retaining the complete clothing region. Subsequently, it optimizes the image generated in the previous stage by addressing issues such as ghosting artifacts and clothing artifacts based on the real model’s skin, facial, and other features, producing high-quality, photorealistic model images. To validate the effectiveness of our proposed Mannequin2Real+, we conduct experiments on the high-resolution DeepFashion dataset. Additionally, we construct a real-world dataset of photographed mannequin models for testing Mannequin2Real+. The experimental results demonstrate the efficacy of our method.