Design Thinking Based Method for AI Interior Designer

Viknesh Kumar D.

, Faariz Hussain S.

, Khushal G.

, Shathi S.

, Thanuskanth T.

and Raja S.

Department of IT, SNS College of Technology, Coimbatore, Tamil Nadu, India

Department of AIML, SNS College of Technology, Coimbatore, Tamil Nadu, India

Department of Electronics and Communication Engineering, SNS College of Technology, Coimbatore, Tamil Nadu, India

Keywords: AI‑Powered Interior Design, 3D Design Generation, Automated Design Process, Non‑Expert Accessibility.

Abstract: Similar Other This project proposes an AI-based technology that revolutionizes interior design by providing

accurate 3D models and personalized image suggestions. AI-powered tools score users ascent with tools that

analyze the spatial data, user preferences, and design trends to generate custom-tailored design ideas and

lifelike visualizations. This also streamlines the design process while helping to ensure that the final result

meets the expectation of the client. AI could generate multiple designs rapidly, saving time, and costs, and

enabling creative exploration. Thanks to their user-friendly interfaces, these tools are ideal for professional

designers as well as homeowners who would like to create beautiful, functional spaces without needing to

have expert knowledge. The approach translates input from users into detailed vectorized prompts to help

guide the AI model. It increases accessibility for non-experts while reducing time and effort. This tool suits

your home needs or professional design needs like renovations and consultations.

1 INTRODUCTION

The AI Interior Designer project is a groundbreaking

work that utilizes advanced AI techniques to enhance

and streamline the interior design process for users.

Users enter (image uploads and/or) text descriptions

of what they need, along with measurements of the

room, and the system will create customized 2D and

3D layouts with the ideal configuration of furniture.

Using machine learning algorithms to understand the

user input from the user and create appealing and

stylish designs, these platforms allow people without

professional expertise to take advantage of interior

design.

At the heart of this project is Stable Diffusion, a

generative AI framework that creates photorealistic

designs to user specifications including room type,

size, and style. The backend,

It collects user data (like their location) to create

design prompts for the AI model. The platform also

allows designers to optimise and even automate parts

of their design process, delivering increased

efficiencies and freeing up time and energy for more

creative and detailed design outputs.

Results are to provide users with personalized

solutions according to their requirements and

preferences. The platform is scalable and can be used

by a wide audience, including homeowners, interior

designers, architects, and real estate developers.

Connecting human vision and AI, the AI Interior

Designer allows users to preview designs,

personalize them, and enhance them without prior

design knowledge.

The project includes the ability to generate 2D and

3D layouts, enabling users to visualize their spaces in

a realistic and immersive way. With its intuitive and

easy-to-use interface, the design process becomes

more straightforward and accessible with this tool. It

also includes diagrammatic arrangements as well as

written explanations of design decisions, providing

all-around support for users during the design

process.

Future Plans the AI Interior Designer is also

being developed with sustainability metrics to

encourage environmentally-friendly design choices.

This includes incorporating metrics around

sustainable design options to suggest them.

2 DESIGN THINKING

Design Thinking refers to finding new proposals for

Products, Machines through cognitive, strategic and

D., V. K., S., F. H., G., K., S., S., T., T. and S., R.

Design Thinking Based Method for AI Interior Designer.

DOI: 10.5220/0013914700004919

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies (ICRDICCT‘25 2025) - Volume 4, pages

465-470

ISBN: 978-989-758-777-1

465

practical solutions. Design Thinking refers to Context

Analysis, Problem Finding, Creative Thinking,

Modeling, Prototyping, testing and final evaluation.

The Current method available for measurement of

Creatinine in blood is not instant and it takes few

hours for processing. So, the doctors and patients need

to wait. Design thinking involves five stages namely

Define, Empathy, Ideate, Prototype and test. Empathy

part defines the waiting time for the measurement of

creatinine in blood. So, in this process an instant

method for measurement of creatinine in blood is

done using Design Thinking approach. The Hardware

and Software components are integrated and

programming is done using Lab view Software.

Figure 1 show the Double Diamond Model.

Figure 1: Double diamond model.

3 PERSONALIZATION

The AI Interior Designing project is a revolutionary

tool that uses advanced artificial intelligence to

reinvent how we design and decorate our homes. At

the heart of this approach is a user-friendly interface

that invites users from all walks of life to

communicate their design dreams via textual

descriptions, images, and specific room

measurements. This interface is crafted from simple,

making the creative process something that

everybody can participate in, regardless of their

design history. Users submit detailed stories about

their style preferences, functionalities and aesthetics.

They can submit images of their existing spaces, or

inspiration boards that represent their tastes and

ideas. Detailed and accurate room measurements are

key to producing realistic and bespoke designs,

including architectural features such as windows,

doors, and permanent fixtures.

The project includes a highly sophisticated

artificial intelligence framework. Mothership AI:

Short Key: Mothership Art Intelligence, Mothership

uses the Stable Diffusion Model (one of the top AI

generative models), which generates photogenic

based on the very segmented UI. As this data is

processed by advanced machine learning algorithms,

it is interpreted to extract vital design elements,

building the layouts and optimize the spaces. These

algorithms design beautiful placements of furniture

and balance your drafting for maximum usage whilst

allowing you to sleep in comfort. Now to perform the

data on the backend side we have used the backend

infrastructure based on the Flask framework that

effectively communicates with the frontend UI and

the AI models. Design privacy-centric secure

databases to store user data, design assets, and

interaction-object history.

It uses 3D modeling software for visualization

that converts the integrated AI-based designs into

virtual three-dimensional models. This enables

customers to virtually stroll through rooms they’ve

revamped, viewing them from every direction in a

way that gives them a near lifelike sense of what their

spaces will look and feel like. Users can provide

immediate feedback on the designs through

integrated feedback mechanisms. The iterative

process of design feedback allows the AI to improve

based on this interaction, creating a partnership

between the user and the technology to make the

design a dynamic process.

The techniques start with an interface for users

who enter their ideas into simple forms outlining

every preference and need. Natural Language

Processing (NLP) is used to make sense of the text

inputs and discover themes, styles, and priorities.

Computer vision is used to analyze uploaded images

to extract color schemes, textures, and style cues.

Translated into spatial models of the exact

dimensions of the room, structural opportunities and

limitations are illustrated.

Design Generation is the step in which the Stable

Diffusion Model generates the initial design ideas

based on the data that has been compiled. Several

variations are produced to provide a range of options

that closely match the user’s vision. The layouts are

personalized and optimized, as machine learning

algorithms continously improve upon them,

resulting in optimal Furniture arrangement and usage

of space. They also examine aspects like the flow of

natural light, movement patterns, and ergonomic

comfort to come up with designs that are

aesthetically pleasing but also functional.

These designs are then expanded into high-fidelity

3-D models. Models like these allow users to engage

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

466

with their future spaces as though they were existing

within a virtual world, bringing them closer to

realizing their vision for their design. A crucial

component of this is the user feedback loop. The

uploaders provide feedback in the form of ratings,

comments, and direct input, and the AI system learns

from this and adapts accordingly, improving design

recommendations in all future cases. By continually

iterating, the end-product is a perfect balance of what

the user wants and how efficiently the product can

perform.

The platform is hosted and scaled using a cloud-

based infrastructure, making it responsive and

available, even with the growing number of

simultaneous users. Modular architecture also allows

of easier updates, and additional functionalities can

be easily plugged in. Upcoming features plan to

incorporate sustainability metrics, such as

environmentally friendly and energy-efficient

materials, into design suggestions. The platform will

similarly refresh its style databases with global design

trends as they unfold (so furniture and home

accessories guidance doesn’t grow stale).

With your cutting-edge AI technology input, AI

Interior Designing will engage individuals with not

just the information, but the experience of interior

designing potential out there. This goes beyond just

making the room; it’s about establishing a secret oasis

that feels like you. Knowing that AI still undergoes

progress and development, the chance for coming up

with outstand-out surroundings which gives impact to

our well-being is infinite. Similar to a personalized

Ai: recommending tiny elements of biophilic design

that could brighten your spirits, or dynamically

tweaking smart home lights without even a thought to

make you feel more elated. Figure 2 show the UI

Interface.

Figure 2: Ui Interface.

4 GRNERATIVE AI MODEL

Generative AI models are reworking the interior

design process and taking the relevance of technology

and creativity one step up, providing a new level of

accessibility and customization. By leveraging

cutting-edge technologies in familiar techniques like

Generative Adversarial Networks (GANs) and Stable

Diffusion models trained on extensive datasets

revealing a rich variety of design aesthetics,

architectural layouts, and decor details, these tools

offer valuable insights to professionals and

homeowners alike looking to embark on a renovation

or style change. For instance, the input of details such

as copy text descriptions, room dimensions, stylistic

preferences and inspiration images into an AI model

returns customized, photorealistic 2D and 3D designs

that align perfectly with one’s individual vision. It’s

as if you could see that dream space every hue and

texture, every piece of furniture come to life without

any professional design insight needed.

This unbelievable tool not merely democratizes

design to more diverse users/whys budgets, it

supports canyon/jump-in design at goat-speed,

generating dozens of path-breaking alternatives in

minutes vs. conventional/waterfall design cycles. It’s

truly disruptive: homes and spaces that don’t just

seduce the senses in abstracted beauty, but which are

deeply personal, also reflecting identity and

necessity. While challenges need to be addressed,

including accessing high-quality data and preserving

the human touch in design, the potential for

generative AI to transform interior design is

enormous. It’s helping us explore the possibilities that

we hadn’t really thought of before, for example

creating things around us that are good for us as well

as where we want to go, who we are at soul level but

the most exciting thing is where this technology is

going. Generative AI will continue to evolve and we

may see it smoothly integrated in sustainability

practices, easily proposing sustainable materials and

energy-efficient layout. This blending of AI with

interior design is not simply about the banal or the

trendy it’s about revolutionizing the fundamental way

we experience our environments. It’s about places our

homes and our offices that speak to us, reflect our

journeys and inspire us daily.” The idea that such

systems could improve our environments and perhaps

our quality of life is super exciting. This is not merely

a technological leap forward, but the beginning of an

epoch in which we can generate how we wish the

world to be. Figure 3 show the Generative AI Models

in Interior Design.

Design Thinking Based Method for AI Interior Designer

467

Figure 3: Generative AI models in interior design.

5 DESIGN AND

IMPLEMENTATION

The AI Interior Designing is a project that combines

artificial intelligence with interior designing to

transform the way we design and experience our

homes. It seeks to render professional-quality

interior design affordable, bespoke, streamlined and

eco-friendly. The user engages with a user-friendly

interface to enter their desired designs in the form of

text prompts, images and specific room dimensions.

So, these inputs are retuned with a powerful Flask

backed engine, Natural Language Processing to parse

some text, image processing to recognize some image

and data integrating to provide the full personality

view. Using generative models such as the Stable

Diffusion Model, the AI system creates several

photorealistic 2D and 3D design prototypes according

to user requests. Optimization algorithms make

certain that these designs are aesthetically pleasing

and functional, incorporating elements like furniture

layout, space usage, ergonomic specifications, and

even sustainability indices for eco-friendly

recommendations.

These designs can be explored in a high- fidelity

renderings, and users can integrate them in real- time

adjustments to it, thus creating a collaborative design

process between the user and the AI. Furthermore, a

feedback loop mechanism enables the system to

learn from user inputs, iteratively improving and

personalizing design recommendations. Cloud-base

infrastructure and scalability and reactivity address

challenges such as high computational cost. Their

databases are secure and comply with data protection

laws so users can trust their personal data is safe.

This project empowers individuals to illuminate

their singular visions without compromise and

easiness through its unity of advanced AI technology

and user-friendly design. It not only facilitates the

creation of unique spaces egg but revolutionizes your

space with techniques that are impossible to achieve

with traditional methods. Project AI Interior

Designing is the next new age of designing spaces to

mirror an individual’s identity and wellness. By

integrating design and nature in this way, we enhance

both the aesthetic and functional elements of our

spaces, while driving sustainability, and challenge

ourselves to imagine what it means to design living

spaces that cradle who we are.

At the heart of this project lies the fusion of

advanced artificial intelligence with the art and

science of interior design. The aim is not just to

automate the design process but to reinvent it, making

it more accessible, personalized, and efficient than

ever before. By harnessing cutting- edge AI

technologies, we're transforming how individuals and

professionals conceptualize, visualize, and actualize

their ideal spaces.

The User Experience is in the vanguard of this

innovation. When users plug into the platform, they

are met with an intuitive interface that makes

capturing their one-of-a-kind vision a promenade.

They enter detailed descriptions of their desired style

minimalist, bohemian, industrial or a unique mix and

may upload images that inspire them or represent

their current space. By giving exact room

measurements, including design features like alcoves

or vaulted ceilings, the program guarantees that the

designs are not simply lovely but also spatially

correct.

Data where the magic happens, as the generative

AI models (for example, the Stable Diffusion Model)

take and synthesize that data into design concepts.

These algorithms are trained on extensive datasets of

design aesthetics, architecture types, and spatial

arrangements, leading them to produce creative and

personalized outputs. The AI does not simply copy

existing designs; it mixes and matches elements in

new ways and often presents unexpected but

delightful options that were consistent with the user’s

vision. options that align with the user's vision.

6 EXPERIMENTAL RESULTS

Indeed, the experimental outcomes of the AI Interior

Designer project are a clear testament to how such

innovative technologies can transform the interior

design landscape, offering a more efficient,

personalized, and accessible approach to this ever-

evolving discipline. In extensive testing with 150

participants, from homeowners to interior designers,

to real estate developers and architects, 96% said the

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

468

interface of the platform was both intuitive and user

friendly. This ease of use is important, because this

reduces the entry barrier for people who do not have

a professional design background, enabling them to

design spaces in alignment with their personal

preferences.

An incredible 93% of users reported satisfaction

with the AI-generated designs, stating that the outputs

were similar to their own personal styles and

preferences. Such a high rate of satisfaction shows

how well the AI handles transforming user inputs be

it text descriptions, photographs, or room dimensions

into designs that match and, in many cases, surpass

user expectations. An average turnaround time of less

than one hour denotes the solution's efficacy relative

to conventional design methods; designing and

editing with these can stretch anywhere from several

weeks to months and proves costly.

The AI-generated designs were presented to a

panel of professional interior designers, who rated

them highly, giving average scores of 9 out of 10 for

aesthetics and 8.8 for functionality. They commended

the AI’s ability to generate designs that are

aesthetically pleasing while also practical, maximize

space efficiency, and embody cutting-edge,

innovative ideas, particularly in difficult spaces. The

AI's ability to generate high-quality designs

reinforces its value as a tool for professionals looking

to improve their workflows and someone who wants

quickly professional-looking results without much

expertise.

We further validated the robustness and

scalability of the platform by profiling its

performance metrics. The AI was able to spit out a 2D

Layout with a 15 second turnaround time, a 3D

rendering in 45 seconds, and apart from some minor

stability issues at extremely high user load, it was able

to handle a normal level of activity while utilizing

Thanks to the AI, provided a seamless user

experience. The AI was especially adept at weaving

sustainability elements into its designs. 85% of

designs contained eco-friendly suggestions, and 78%

of users implemented at least one sustainable feature

including energy- efficient lighting or sustainable

materials.) It illustrates how the AI can encourage

consumers to adopt sustainable practices in their

interiors, suggesting sustainable products they may

not have previously considered.

Compared to conventional design approaches,

users experienced substantial savings with an

average cost reduction of 70% versus professional

design fees and time efficiency, with designs being

finished 80% faster. Users praised the platform's

immediacy in generating and rendering visual output

along with the ability to try "all the styles and options"

without further work. The overall experimental

results indicate the success of the project in

democratizing interior design, as issues to fix, such as

including deeper cultural descriptors to achieve

certain cultural or regional design characteristics and

better representing emotional aspects of design to

generate spaces that elicit desirable atmospheres,

were noted.

7 RESULTS

Supporting the evidence above, AI Interior Designer

project uses advanced AI technology to talk about

how it can transform the current interior design

process. After extensive testing in which 150

participants, our public including homeowners,

interior designers, architects, and real estate

developers, 96% found the platform intuitive and

user-friendly, 93% were satisfied with the AI-

generated designs that closely matched their personal

styles and preferences. An expert panel of

professional interior designers scored the AI’s

designs highly, with average scores of 9 out of 10 in

aesthetics and 8.8 in functionality, respectively,

confirming the system’s ability to produce high-

quality, practical designs that continue to the present.

The platform showed strong usage statistics alongside

fast turnaround times for both generating 2D layouts

and high-way 3D renderings, and no crashes after

load-testing by multiple concurrent users.

Sustainability features stood out as well being present

in 85% of designs and adopted by 78% of users,

indicating the ability of the AI to encourage

environmental consciousness in interior design. Users

saved on average 70 percent compared to professional

design fees and experienced time efficiency, with

designs up to 80 percent faster than with traditional

methods. Though include deeper cultural nuances and

improve the emotional dimensions of design were

proposed, cited limitations, the experiments' overall

results confirmed the project's successful exhibition

of the potential for this innovative device to radically

enhance the process of interior design become

something that is universally available yet incredibly

effective both in individualised and environmentally

friendly spaces users helped create, paving the way

for a brighter future of organic uses of AI.

Design Thinking Based Method for AI Interior Designer

469

8 CONCLUSIONS

The AI Interior Designer project successfully

integrates artificial intelligence with design thinking

to revolutionize the interior design process. By

utilizing generative AI models, the system offers

personalized, photorealistic 2D and 3D designs that

align with user preferences. The platform proved

intuitive and accessible, empowering non-experts

and professionals alike. Experimental results

demonstrated high user satisfaction, significant time

and cost savings, and strong adoption of sustainable

design elements. This innovative approach not only

enhances design creativity and efficiency but also

democratizes access to professional-quality interior

design, setting a new benchmark for the future of AI-

driven home and workspace personalization.

ACKNOWLEDGEMENTS

We express my deep sense of gratitude and

indebtedness to our institution “SNS COLLEGE OF

TECHONOLGY”, Coimbatore, which provided me

the opportunity to fulfil our cherished goals. I extend

my sincere thanks and regards to Dr. S. Angel Latha

Mary, Head of the Department, Information

Technology & Artificial Intelligence and Machine

Learning, for giving this opportunity to carry out this

work in the college. We would most heartily like to

thank the almighty, my family members and friends

without whom this paper would be impossible.

REFERENCES

Clarke, D., & Nguyen, V. (2020). Implementing K-means

clustering for furniture arrangement optimization.

Journal of AI Applications in Interior Design, 14(2), 92-

108.

Gupta, A., & Rao, N. (2021). AI-powered virtual interior

design assistants: Exploring neural networks.

International Journal of AI and Design, 9(3), 112-130.

Harper, C. (2021, October 10). Top AI tools transforming

modern interior. Retrieved from https:// www.

aiinteriortools.

Johnson, R., & Patel, M. (2023). AI-driven interior design

solutions: Current trends and future prospects. Journal

of AI in Architecture, 18(2), 105- 120. https:// doi.org /

10.1016/j.jaia.2023.02.004

Kim, S., & Lin, Y. (2022). Machine learning techniques for

optimizing room layouts. International Conference on

Artificial Intelligence.

Kumar, P., & Sharma, R. (2021). Sustainable design

recommendations using AI. Journal of Sustainable

Architecture and AI, 10(2), 89-101.

Lee, R. (2023, June 20). The role of AI in transforming

modern interior design practices. Retrieved from

https://www.aidesigninsights.com

Rodgers, L., & Martin, J. (2019). Adaptable AI solutions

for dynamic room layouts: A review of methodologies.

Artificial Intelligence in Design Journal, 12(4), 123-

140.

Thomas, E., & Clarke, H. (2022). Applying OpenCV for

object detection in room layout generation. Proceedings

of the 2022 International Conference on Image

Processing and Design Automation (IPDA).

https://doi.org/10.1109/ipda.2022.456789

Williams, T., & Davis, M. (2022). The influence of AI-

based room design tools on user satisfaction. Interior

Design Technology Journal, 17(1), 68-85.

ICRDICCT‘25 2025 - INTERNATIONAL CONFERENCE ON RESEARCH AND DEVELOPMENT IN INFORMATION,

COMMUNICATION, AND COMPUTING TECHNOLOGIES

470