Academia.eduAcademia.edu

Figure 12 – uploaded by Hanhui Deng

See full PDFdownloadDownload figure
tyleMe — (b) The Scores from Design Group (c) The Scores from Non-Design Group (d) The 7, for Each Pair of Works

Figure 12 tyleMe — (b) The Scores from Design Group (c) The Scores from Non-Design Group (d) The 7, for Each Pair of Works

Related Figures (13)

Figure 1: StyleMe system employs generative adversarial network to automatically generate personalized clothing sketches based on designer’s sketches and transfer style from sketch to real clothing image. This figure exhibits the sketch images and clothing images generated by the system. The code, dataset and video demonstration of StyleMe are available at following project link: https://github.com/ExponentiAI/StyleMe. ABSTRACT clothes, especially when the fashion design requires distinctive and personal characteristics from designer style. This paper presents an artificial intelligent aided fashion design system, namely StyleMe, to support the automatic generation of clothing sketches with de- signer style. Given the clothing pictures specified by the designer, StyleMe can use deep learning based generative model to generate clothing sketches that are consistent with the designer style. The system also supports intelligent colourization on clothing sketch by style transfer, according to specified styles from the real fashion im- ages. Through a series of performance evaluations and user studies, we found that our system can generate effective clothing sketches as good as fashion designers’ human work, and significantly im- prove the efficiency of fashion design with its sketch colourization method. Hand-drawn sketches and sketch colourization are the most labori- ous but necessary steps for fashion designers to design exquisite
Figure 1: StyleMe system employs generative adversarial network to automatically generate personalized clothing sketches based on designer’s sketches and transfer style from sketch to real clothing image. This figure exhibits the sketch images and clothing images generated by the system. The code, dataset and video demonstration of StyleMe are available at following project link: https://github.com/ExponentiAI/StyleMe. ABSTRACT clothes, especially when the fashion design requires distinctive and personal characteristics from designer style. This paper presents an artificial intelligent aided fashion design system, namely StyleMe, to support the automatic generation of clothing sketches with de- signer style. Given the clothing pictures specified by the designer, StyleMe can use deep learning based generative model to generate clothing sketches that are consistent with the designer style. The system also supports intelligent colourization on clothing sketch by style transfer, according to specified styles from the real fashion im- ages. Through a series of performance evaluations and user studies, we found that our system can generate effective clothing sketches as good as fashion designers’ human work, and significantly im- prove the efficiency of fashion design with its sketch colourization method. Hand-drawn sketches and sketch colourization are the most labori- ous but necessary steps for fashion designers to design exquisite
Figure 2: The user interface of StyleMe system consists of four parts, (1) inspiration pictures area is used for users to browse clothing pictures and select pictures to generate sketches; (2) drawing board area is used to show the generated sketch from selected clothing picture, and the designer can modify the sketch; (3) reference pictures area is used for users to browse clothing style and select pictures to transfer style; (4) exhibition board area is used to display a list of clothing pictures after style transfer.
Figure 2: The user interface of StyleMe system consists of four parts, (1) inspiration pictures area is used for users to browse clothing pictures and select pictures to generate sketches; (2) drawing board area is used to show the generated sketch from selected clothing picture, and the designer can modify the sketch; (3) reference pictures area is used for users to browse clothing style and select pictures to transfer style; (4) exhibition board area is used to display a list of clothing pictures after style transfer.
Figure 4: An overview of personalized sketch automatic generation network architecture.
Figure 4: An overview of personalized sketch automatic generation network architecture.
Figure 5: An overview of style transfer generation network architecture.
Figure 5: An overview of style transfer generation network architecture.
Figure 6: An overview of the style transfer pipeline in StyleMe, which inputs a sketch and multiple reference RGB clothing images, and outputs new clothing images with de- signer style and intelligent colourization.
Figure 6: An overview of the style transfer pipeline in StyleMe, which inputs a sketch and multiple reference RGB clothing images, and outputs new clothing images with de- signer style and intelligent colourization.
Table 1: The quantitative comparison to other methods and variants.
Table 1: The quantitative comparison to other methods and variants.
Related topics:
Computer ScienceHuman Computer InteractionFashion design
580 California St., Suite 400
San Francisco, CA, 94104
© 2025 Academia. All rights reserved