Refactoring Web Virtual Try-On Berbasis AI dengan Framework Next.js

Hazdi, Mavaldi Rizqy (2026) Refactoring Web Virtual Try-On Berbasis AI dengan Framework Next.js. Project Report. [s.n.], [s.l.]. (Unpublished)

Text 5025211086-Project_Report.pdf - Accepted Version Restricted to Repository staff only Download (749kB) | Request a copy

Abstract

Perkembangan teknologi belanja online menuntut adanya solusi yang mampu meningkatkan pengalaman pengguna dalam memilih pakaian secara virtual, namun aplikasi web virtual try-on yang telah dikembangkan sebelumnya memerlukan pembaruan agar tetap optimal, aman, dan kompatibel dengan teknologi terkini. Oleh karena itu, kerja praktik ini bertujuan untuk melakukan refactoring terhadap aplikasi web virtual try-on berbasis AI guna meningkatkan performa sistem, keamanan, serta kemudahan akses pengguna. Metode yang digunakan meliputi pembaruan frontend menggunakan framework Next.js, pengembangan backend yang di-deploy melalui Railway, integrasi dan pembaruan model AI dari Hugging Face, serta migrasi dan pengelolaan database serta penyimpanan gambar menggunakan Supabase Storage. Setelah proses refactoring dan integrasi selesai, dilakukan pengujian terhadap fitur utama seperti unggah gambar pengguna, pemilihan dan pengelolaan data pakaian, proses rendering gambar, penyimpanan hasil, serta pengujian deployment sistem. Hasil pengujian menunjukkan bahwa seluruh fitur berjalan dengan baik, proses rendering memiliki waktu rata- xiii rata 45–55 detik dengan tingkat akurasi yang sesuai desain, serta sistem berhasil di-deploy dan diakses secara stabil, sehingga aplikasi mampu memberikan pengalaman virtual try-on yang lebih optimal dan responsif dibandingkan versi sebelumnya.
====================================================================================================================================
Technological developments in online shopping demand solutions capable of enhancing the user experience when selecting clothing virtually; however, previously developed virtual try-on web applications require updates to remain optimal, secure, and compatible with the latest technologies. Therefore, this project aims to refactor an AI-based virtual try-on web application to improve system performance, security, and user accessibility. The methods employed include frontend updates using the Next.js framework, backend development deployed via Railway, integration and updates of AI models from Hugging Face, as well as database migration and management along with image storage using Supabase Storage. After the refactoring and integration processes were completed, testing was conducted on key features such as user image upload, clothing data selection and management, image rendering, result storage, and system deployment testing. Test results showed that all features functioned properly, the rendering process had an average duration of 45–55 seconds with accuracy levels meeting design specifications, and the system was successfully deployed and accessed stably, enabling the application to provide a more optimal and responsive virtual try-on experience compared to the previous version.

Item Type:	Monograph (Project Report)
Uncontrolled Keywords:	Deploy, Hugging Face, Next.js, Refactoring, Supabase, Virtual try-on, meluncurkan, Hugging Face, Next.js, refaktorisasi, Supabase, Virtual try-on.
Subjects:	T Technology > T Technology (General) > T58.8 Productivity. Efficiency
Divisions:	Faculty of Intelligent Electrical and Informatics Technology (ELECTICS) > Informatics Engineering > 55201-(S1) Undergraduate Thesis
Depositing User:	Mavaldi Rizqy Hazdi
Date Deposited:	08 Apr 2026 01:00
Last Modified:	08 Apr 2026 01:00
URI:	http://repository.its.ac.id/id/eprint/132768

Actions (login required)

View Item