Pemilihan Kosmetik Berbasis Warna Kulit Wajah dengan Metode CNN-GAN

Adhelina, Aysha Nabila (2021) Pemilihan Kosmetik Berbasis Warna Kulit Wajah dengan Metode CNN-GAN. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img] Text
07211640000041-Undergraduate_Thesis.pdf - Accepted Version

Download (3MB)

Abstract

Perkembangan teknologi semakin lama maju dan tentunya kemajuan teknologi ini telah mempengaruhi seluruh sektor kehidupan manusia. Dengan kemajuan teknologi ini, kegiatan berbelanja dapat dilakukan secara online, dimana kita tidak perlu lagi pergi keluar untuk berbelanja. Akan tetapi berbelanja secara online juga memiliki beberapa kekurangan seperti tidak bisa mencoba dan melihat barang yang diinginkan secara langsung dan dalam hal kosmetik, pembeli pasti merasa ragu apakah kosmetik yang dibeli cocok dengan pembeli atau tidak terutama dalam segi kecocokan dengan warna kulit pembeli. Dari permasalahan tersebut, diperlukannya suatu sistem untuk membantu para pembeli kosmetik apakah kosmetik yang akan mereka beli akan sesuai dengan warna kulit pembeli atau tidak. Dengan menggunakan metode CNN-GAN dibuatlah sistem rekomendasi yang akan merekomendasikan produk kosmetik dan riasan wajah (makeup) yang cocok atau sesuai dengan warna kulit wajah pengguna yang akan dimasukkan ke dalam sistem. Hasil pengujian menunjukkan bahwa hasil rekomendasi produk kosmetik dan riasan wajah sudah sesuai dengan warna kulit wajah sebesar 60.925%. Dan berdasarkan kuisioner terhadap 132 responden, 55.07% responden setuju bahwa riasan wajah yang direkomendasikan sesuai dengan warna kulit wajah, dan 54.8267% responden setuju bahwa produk kosmetik yang direkomendasikan sesuai dengan warna kulit wajah. ======================================================================================================== The development of technology has progressed over time and of course this technological progress has affected all sectors of human life. With this technological advancement, shopping activities can be done online, where we no longer need to go out to shop. However, online shopping also has some drawbacks such as not being able to try and see the desired item in person and in the case of cosmetics, buyers must have doubts whether the cosmetics purchased are suitable for the buyer or not, especially in terms of matching the buyer's skin color. From these problems, we need a system to help cosmetic buyers whether the cosmetics they will buy will match the buyer's skin color or not. Using the CNN-GAN method, a recommendation system is created that will recommend cosmetic products and makeup that match for the user's facial skin color. The test results show that the results of the recommendations for cosmetic and facial makeup products are in accordance with facial skin color by 60,925%. And based on the questionnaire of 132 respondents, 55.07% of respondents agreed that the recommended facial makeup was in accordance with the skin color of the face, and 54.8267% of the respondents agreed that the recommended cosmetic products match the skin color of the face.

Item Type: Thesis (Undergraduate)
Uncontrolled Keywords: Image to Image Translation, Generative Adversarial Network, Makeup, Facial Makeup Transfer.
Subjects: Q Science > QA Mathematics > QA76.87 Neural networks (Computer Science)
T Technology > T Technology (General)
T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques
Divisions: Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Computer Engineering > 90243-(S1) Undergraduate Thesis
Depositing User: Aysha Nabila Adhelina
Date Deposited: 10 Mar 2021 03:57
Last Modified: 10 Mar 2021 03:57
URI: https://repository.its.ac.id/id/eprint/83959

Actions (login required)

View Item View Item