Accesso libero

Research on Traditional Costume Culture Information Extraction and Digital Reconstruction Methods Based on Artificial Intelligence

Miao Yu

Miao Yu

Faculty of Art and Design, Shanghai Business SchoolShanghai, China
Cerca questo autore su
Sciendo|Google Scholar
Yu, Miao
  
24 mar 2025
Research on Traditional Costume Culture Information Extraction and Digital Reconstruction Methods Based on Artificial Intelligence's Cover Image
INFORMAZIONI SU QUESTO ARTICOLO
Articolo precedente
Articolo Successivo
Cita
Scarica la copertina
Pubblicato online: 24 mar 2025
Ricevuto: 04 nov 2024
Accettato: 03 feb 2025
DOI: https://doi.org/10.2478/amns-2025-0718
Parole chiave
© 2025 Miao Yu, published by Sciendo
This work is licensed under the Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Figure 1.

The training curve and the validation curve of the RSKP-UNet model
The training curve and the validation curve of the RSKP-UNet model

Figure 2.

Model training loss curve of different learning rate
Model training loss curve of different learning rate

Figure 3.

Model training loss curve of different batch size
Model training loss curve of different batch size

Figure 4.

Evaluation score of different professional respondents
Evaluation score of different professional respondents

Information extraction results of the UNet model

Site AP Precision F1 value Recall
Arm 72.76% 78.29% 77.84% 75.99%
Foot 62.34% 70.27% 70.22% 70.46%
Forebreast 78.47% 77.35% 83.96% 81.31%
Head 92.46% 90.14% 90.39% 91.58%
Leg 68.17% 63.09% 70.92% 73.87%
Waist 70.17% 67.58% 65.33% 72.49%

Similarity of different element images

Pattern elements of clothing SS MSE MSRE PSNR
Elephant, pagoda 0.689 0.775 0.851 18.693
Elephant, human 0.684 0.769 0.859 17.857
Peacock 0.779 0.925 0.946 20.672
Human, horse 0.725 0.828 0.919 18.683
Architecture 0.846 0.781 0.861 17.657
Average 0.745 0.816 0.887 18.712

Information extraction results of the RSKP-UNet model

Site AP Precision F1 value Recall
Arm 81.88% 81.05% 78.74% 80.67%
Foot 73.42% 74.05% 75.52% 72.55%
Forebreast 84.47% 86.74% 85.04% 85.88%
Head 95.62% 93.29% 92.71% 94.33%
Leg 71.54% 68.94% 72.17% 75.24%
Waist 74.23% 73.03% 70.18% 79.74%
Lingua:
Inglese
Frequenza di pubblicazione:
1 volte all'anno
Argomenti della rivista:
Scienze biologiche, Scienze della vita, altro, Matematica, Matematica applicata, Matematica generale, Fisica, Fisica, altro
Feed RSS della rivista