Wardana, Ahmad Aditya (2023) Analisa Perbandingan Antara Textual Inversion dan Lora Dreambooth pada Stable Diffusion untuk Training Pengenalan Fitur pada Pembuatan Gambar Berbasis AI. Undergraduate thesis, Fakultas Teknologi Informasi Universitas Merdeka Malang.

	Text COVER.pdf Download (1MB)
	Text BAB I.pdf Download (2MB)
	Text BAB II.pdf Restricted to Repository staff only Download (2MB)
	Text BAB III.pdf Restricted to Repository staff only Download (2MB)
	Text BAB IV.pdf Restricted to Repository staff only Download (2MB)
	Text BAB V.pdf Restricted to Repository staff only Download (2MB)
	Text DAFTAR PUSTAKA.pdf Download (2MB)
	Text HASIL CEK PLAGIASI.pdf Download (2MB)

Abstract

Artificial intelligence (AI) or artificial intelligence is intelligence that is added to a system or in other words the ability of the system to properly hide external data and manage this data and use the processed results for a particular purpose. One of the features of Artificial intelligence is ability to perform image generation. Image Generation is the process of automatically generating new images using computer technology, Text-to-image generation refers to creating visually realistic images, progress has been made in producing visually realistic images. In this research study, we want to compare Textual Inversion with LoRa (low-rank adaptation) which is used in facial recognition artificial intelligence. This test aims to find out which one is more efficient in terms of similarity, training time and file size. Using a comparative research method with a quantitative approach, and using a Likert scale involving 30 respondents. The test results show that LoRa is superior in similarity training, in terms of size Textual is more efficient although not significant and in terms of training speed LoRa training is more efficient.

Actions (login required)

View Item