Enhancing interior design with AI-enhanced prompts and stable diffusion
Traditional interior design CAD tools often fall short in terms of user-friendliness and inclusivity, leading to user frustration and dissatisfaction. In response, RoomVision AI leverages advanced Generative AI technology to transform the landscape of interior design. Users can input either a sketch...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Final Year Project |
Language: | English |
Published: |
Nanyang Technological University
2023
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/171950 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
id |
sg-ntu-dr.10356-171950 |
---|---|
record_format |
dspace |
spelling |
sg-ntu-dr.10356-1719502023-11-17T15:38:01Z Enhancing interior design with AI-enhanced prompts and stable diffusion Sainani, Evan Sean Lin Weisi School of Computer Science and Engineering WSLin@ntu.edu.sg Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Computer science and engineering::Computer applications::Computer-aided engineering Traditional interior design CAD tools often fall short in terms of user-friendliness and inclusivity, leading to user frustration and dissatisfaction. In response, RoomVision AI leverages advanced Generative AI technology to transform the landscape of interior design. Users can input either a sketch or image of an interior alongside a textual description of their aesthetic preferences. This user-centric web application bridges the communication gap between users and CAD tools through natural language refinement using ChatGPT 3.5 Turbo. By integrating these refined prompts into the Stable Diffusion model, users can effortlessly explore interior design concepts. For a personalized experience, ControlNet enhances the Stable Diffusion model, maintaining fidelity to reference images and ensuring generated concepts align with the input structure. RoomVision AI empowers users of all proficiency levels, promising an inclusive and creative future for interior design. The advantages include ease of use, realistic and personalized result generation, and a way to conceptualize a user’s design preferences that can be used for their interior design process. The results of the user feedback survey unequivocally affirm RoomVision AI's advantages, with the average response indicating strong agreement on the web application's user-centric design and satisfaction with the generated concepts. The web application signifies a significant step towards redefining interior design practices for a more inclusive and creative future. Bachelor of Engineering (Computer Science) 2023-11-17T04:27:42Z 2023-11-17T04:27:42Z 2023 Final Year Project (FYP) Sainani, E. S. (2023). Enhancing interior design with AI-enhanced prompts and stable diffusion. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/171950 https://hdl.handle.net/10356/171950 en SCSE22-0805 application/pdf Nanyang Technological University |
institution |
Nanyang Technological University |
building |
NTU Library |
continent |
Asia |
country |
Singapore Singapore |
content_provider |
NTU Library |
collection |
DR-NTU |
language |
English |
topic |
Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Computer science and engineering::Computer applications::Computer-aided engineering |
spellingShingle |
Engineering::Computer science and engineering::Computing methodologies::Artificial intelligence Engineering::Computer science and engineering::Computer applications::Computer-aided engineering Sainani, Evan Sean Enhancing interior design with AI-enhanced prompts and stable diffusion |
description |
Traditional interior design CAD tools often fall short in terms of user-friendliness and inclusivity, leading to user frustration and dissatisfaction. In response, RoomVision AI leverages advanced Generative AI technology to transform the landscape of interior design. Users can input either a sketch or image of an interior alongside a textual description of their aesthetic preferences. This user-centric web application bridges the communication gap between users and CAD tools through natural language refinement using ChatGPT 3.5 Turbo. By integrating these refined prompts into the Stable Diffusion model, users can effortlessly explore interior design concepts. For a personalized experience, ControlNet enhances the Stable Diffusion model, maintaining fidelity to reference images and ensuring generated concepts align with the input structure.
RoomVision AI empowers users of all proficiency levels, promising an inclusive and creative future for interior design. The advantages include ease of use, realistic and personalized result generation, and a way to conceptualize a user’s design preferences that can be used for their interior design process. The results of the user feedback survey unequivocally affirm RoomVision AI's advantages, with the average response indicating strong agreement on the web application's user-centric design and satisfaction with the generated concepts. The web application signifies a significant step towards redefining interior design practices for a more inclusive and creative future. |
author2 |
Lin Weisi |
author_facet |
Lin Weisi Sainani, Evan Sean |
format |
Final Year Project |
author |
Sainani, Evan Sean |
author_sort |
Sainani, Evan Sean |
title |
Enhancing interior design with AI-enhanced prompts and stable diffusion |
title_short |
Enhancing interior design with AI-enhanced prompts and stable diffusion |
title_full |
Enhancing interior design with AI-enhanced prompts and stable diffusion |
title_fullStr |
Enhancing interior design with AI-enhanced prompts and stable diffusion |
title_full_unstemmed |
Enhancing interior design with AI-enhanced prompts and stable diffusion |
title_sort |
enhancing interior design with ai-enhanced prompts and stable diffusion |
publisher |
Nanyang Technological University |
publishDate |
2023 |
url |
https://hdl.handle.net/10356/171950 |
_version_ |
1783955594609164288 |