• Uday Kumar Dosanapudi
• Kambalapalli Venkata Bhanu Prakash
• Gundipudi Dhanu Sai Srinivas
• Kotla Eswar
• Kumbha Venkata Bharath

This paper proposes a novel AI-powered 3D construction modeling tool that integrates diffusion and Low-Rank Adaptation (LoRA) models to generate realistic and customizable 3D room images based on user-specified requirements. The system accepts multimodal input, including textual descriptions, sketches, and mood boards, enabling users to specify layout, aesthetics, and furniture details. Diffusion models are employed to generate initial room layouts and textures, while LoRA models facilitate fine-grained customization through an iterative refinement process based on user feedback. The final output is a high-quality, photorealistic 3D room visualization. By combining the strengths of diffusion and LoRA models, this approach streamlines architectural design workflows, enhances creativity, and democratizes 3D modeling for non-professionals. The proposed tool addresses limitations in existing systems, such as extensive manual input, high technical expertise requirements, and restricted customization options.

Diffusion Models, Low-Rank Adaptation (LoRA), 3D Room Modeling, Multimodal Input, Photorealistic Visualization, AI-Powered Design, Iterative Refinement, Democratization of 3D Modeling

"AI-DRIVEN DIFFUSION AND LORA MODELS FOR CUSTOMIZABLE 3D ROOM VISUALIZATION AND DESIGN ENHANCEMENT", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 3, page no.c596-c611, March-2025, Available :http://www.jetir.org/papers/JETIR2503274.pdf

"AI-DRIVEN DIFFUSION AND LORA MODELS FOR CUSTOMIZABLE 3D ROOM VISUALIZATION AND DESIGN ENHANCEMENT", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 3, page no. ppc596-c611, March-2025, Available at : http://www.jetir.org/papers/JETIR2503274.pdf