Customify: An AI-Powered Interior Design Platform for Photo-Based Room Redesign, Itemised Cost Estimation, and 2D Blueprint Generation | IJCT Volume 13 – Issue 5 | IJCT-V13I5P51
IJCT
International Journal of Computer Techniques
ISSN 2394-2231 · Peer-Reviewed · Open Access
📚 Volume 13, Issue 5
📅 October 15, 2026
📄 Pages 432–438
🔖 ID: IJCT-V13I5P51
Customify: An AI-Powered Interior Design Platform for Photo-Based Room Redesign, Itemised Cost Estimation, and 2D Blueprint Generation
Author(s)
Ch Koushik, K Surya Pavan, N Sai Ganesh, B Yuvaraj
Abstract
Redecorating a home is costly, time-consuming and hard to follow for the average family: to preview a new look, the owner must engage a designer, and then separately price the furnishings and hunt for suitable products. We introduce Customify, a full stack web application that takes one photograph of a room and produces (i) a redesign preview conditioned on a chosen style, (ii) an item-by-item Indian Rupee cost estimate over three budget bands, (iii) hand-picked buying links on Indian online marketplaces, and (iv) a floor-plan blueprint that can be downloaded as a PDF. The system is split into a React.js 18 single-page front end, a FastAPI service layer protected by bcrypt-hashed passwords and JSON Web Tokens, a SQLAlchemy data layer (SQLite for development, PostgreSQL for production), and four domain modules covering image generation, costing, product recommendation and blueprint drawing. Images are produced with Stable Diffusion img2img via the Replicate API whenever a token is available; without one, a deterministic colour-grading routine takes over, so the application keeps working offline. Testing the implementation, we measured a mean end-to-end response time of 155 ms for 1080p uploads and 443 ms for 4K uploads, a mean blueprint rendering time of 36 ms for PDFs of roughly 3.6 KB, and correct behaviour in every authentication and validation check. The price engine holds exact entries for 35 of 35 furniture classes under seven style multipliers. Two shortcomings surfaced during testing; chiefly, the heuristic room-type detector returns “living room” in every case unless the user provides a hint.
Keywords
Interior Design, Generative AI, Stable Diffusion, Image-to-Image Translation, Cost Estimation, Floor-Plan Gen eration, FastAPI, React.js, JSON Web Tokens, Recommendation Systems
Conclusion
We have described Customify, a full-stack platform that turns a room photograph and a style selection into a redesign pre view, an itemised three-tier INR estimate, purchase links and a scaled 2D blueprint, all behind a protected and documented API. The implementation is light, containerised and robust: image generation switches to an offline mode on its own, and measured response time stays under 0.6 s even for 4K inputs. Testing also uncovered concrete faults, above all the room-type fallback bug and the randomly drawn furniture, and these set the near-term roadmap.
References
[1] J. Ho, A. Jain, and P. Abbeel, “Denoising diffusion probabilistic models,” in Advances in Neural Information Processing Systems, vol. 33, 2020, pp. 6840–6851.
[2] R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High-resolution image synthesis with latent diffusion models,” in Proc. IEEE/CVF CVPR, New Orleans, LA, USA, June 2022, pp. 10684–10695.
[3] C. Meng et al., “SDEdit: Guided image synthesis and editing with stochastic differential equations,” in Proc. ICLR, 2022.
[4] L. Zhang, A. Rao, and M. Agrawala, “Adding conditional control to text-to-image diffusion models,” in Proc. IEEE/CVF ICCV, Paris, France, Oct. 2023, pp. 3836–3847.
[5] T. Brooks, A. Holynski, and A. A. Efros, “InstructPix2Pix: Learning to follow image editing instructions,” in Proc. IEEE/CVF CVPR, Vancouver, Canada, June 2023, pp. 18392–18402.
[6] A. Radford et al., “Learning transferable visual models from natural language supervision,” in Proc. ICML, PMLR vol. 139, July 2021, pp. 8748–8763.
[7] P. Dhariwal and A. Nichol, “Diffusion models beat GANs on image synthesis,” in Advances in Neural Information Processing Systems, vol. 34, 2021, pp. 8780–8794.
[8] B. Zhou, A. Lapedriza, A. Khosla, A. Oliva, and A. Torralba, “Places: A 10 million image database for scene recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 40, no. 6, pp. 1452–1464, June 2018.
[9] B. Zhou et al., “Semantic understanding of scenes through the ADE20Kdataset,” Int. J. Comput. Vis., vol. 127, no. 3, pp. 302–321, Mar. 2019.
[10] S. Song, S. P. Lichtenberg, and J. Xiao, “SUN RGB-D: A RGB-D scene understanding benchmark suite,” in Proc. IEEE CVPR, Boston, MA, USA, June 2015, pp. 567–576.
[11] I. Armeni et al., “3D semantic parsing of large-scale indoor spaces,” in Proc. IEEE CVPR, Las Vegas, NV, USA, June 2016, pp. 1534–1543.
[12] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proc. IEEE CVPR, Las Vegas, NV, USA, June 2016, pp. 779–788.
[13] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE CVPR, Las Vegas, NV, USA, June 2016, pp. 770–778.
[14] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Im age quality assessment: From error visibility to structural similarity,” IEEE Trans. Image Process., vol. 13, no. 4, pp. 600–612, Apr. 2004.
[15] F. Ricci, L. Rokach, and B. Shapira, Eds., Recommender Systems Handbook. New York, NY, USA: Springer, 2011.
[16] C. C. Aggarwal, Recommender Systems: The Textbook. Cham, Switzerland: Springer, 2016.
[17] M. Jones, J. Bradley, and N. Sakimura, “JSON Web Token (JWT),” RFC7519, IETF, May 2015.
[18] N. Provos and D. Mazi`eres, “A future-adaptable password scheme,” in Proc. USENIX Annual Technical Conf. (FREENIX Track), Monterey, CA, USA, June 1999, pp. 81–91.
[19] R. T. Fielding, “Architectural styles and the design of network-based software architectures,” Ph.D. dissertation, Univ. California, Irvine, CA, USA, 2000.
[20] M. Bayer, “SQLAlchemy,” in The Architecture of Open Source Appli cations, Vol. II, A. Brown and G. Wilson, Eds. 2012.
[2] R. Rombach, A. Blattmann, D. Lorenz, P. Esser, and B. Ommer, “High-resolution image synthesis with latent diffusion models,” in Proc. IEEE/CVF CVPR, New Orleans, LA, USA, June 2022, pp. 10684–10695.
[3] C. Meng et al., “SDEdit: Guided image synthesis and editing with stochastic differential equations,” in Proc. ICLR, 2022.
[4] L. Zhang, A. Rao, and M. Agrawala, “Adding conditional control to text-to-image diffusion models,” in Proc. IEEE/CVF ICCV, Paris, France, Oct. 2023, pp. 3836–3847.
[5] T. Brooks, A. Holynski, and A. A. Efros, “InstructPix2Pix: Learning to follow image editing instructions,” in Proc. IEEE/CVF CVPR, Vancouver, Canada, June 2023, pp. 18392–18402.
[6] A. Radford et al., “Learning transferable visual models from natural language supervision,” in Proc. ICML, PMLR vol. 139, July 2021, pp. 8748–8763.
[7] P. Dhariwal and A. Nichol, “Diffusion models beat GANs on image synthesis,” in Advances in Neural Information Processing Systems, vol. 34, 2021, pp. 8780–8794.
[8] B. Zhou, A. Lapedriza, A. Khosla, A. Oliva, and A. Torralba, “Places: A 10 million image database for scene recognition,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 40, no. 6, pp. 1452–1464, June 2018.
[9] B. Zhou et al., “Semantic understanding of scenes through the ADE20Kdataset,” Int. J. Comput. Vis., vol. 127, no. 3, pp. 302–321, Mar. 2019.
[10] S. Song, S. P. Lichtenberg, and J. Xiao, “SUN RGB-D: A RGB-D scene understanding benchmark suite,” in Proc. IEEE CVPR, Boston, MA, USA, June 2015, pp. 567–576.
[11] I. Armeni et al., “3D semantic parsing of large-scale indoor spaces,” in Proc. IEEE CVPR, Las Vegas, NV, USA, June 2016, pp. 1534–1543.
[12] J. Redmon, S. Divvala, R. Girshick, and A. Farhadi, “You only look once: Unified, real-time object detection,” in Proc. IEEE CVPR, Las Vegas, NV, USA, June 2016, pp. 779–788.
[13] K. He, X. Zhang, S. Ren, and J. Sun, “Deep residual learning for image recognition,” in Proc. IEEE CVPR, Las Vegas, NV, USA, June 2016, pp. 770–778.
[14] Z. Wang, A. C. Bovik, H. R. Sheikh, and E. P. Simoncelli, “Im age quality assessment: From error visibility to structural similarity,” IEEE Trans. Image Process., vol. 13, no. 4, pp. 600–612, Apr. 2004.
[15] F. Ricci, L. Rokach, and B. Shapira, Eds., Recommender Systems Handbook. New York, NY, USA: Springer, 2011.
[16] C. C. Aggarwal, Recommender Systems: The Textbook. Cham, Switzerland: Springer, 2016.
[17] M. Jones, J. Bradley, and N. Sakimura, “JSON Web Token (JWT),” RFC7519, IETF, May 2015.
[18] N. Provos and D. Mazi`eres, “A future-adaptable password scheme,” in Proc. USENIX Annual Technical Conf. (FREENIX Track), Monterey, CA, USA, June 1999, pp. 81–91.
[19] R. T. Fielding, “Architectural styles and the design of network-based software architectures,” Ph.D. dissertation, Univ. California, Irvine, CA, USA, 2000.
[20] M. Bayer, “SQLAlchemy,” in The Architecture of Open Source Appli cations, Vol. II, A. Brown and G. Wilson, Eds. 2012.
📋 How to Cite This Paper
Ch Koushik, K Surya Pavan, N Sai Ganesh, B Yuvaraj (2026). Customify: An AI-Powered Interior Design Platform for Photo-Based Room Redesign, Itemised Cost Estimation, and 2D Blueprint Generation. International Journal of Computer Techniques, 13(5), 432–438. ISSN: 2394-2231. DOI: https://doi.org/10.5281/zenodo.23388647
Related Posts:
