Authors : Nithin G.; Pramatha K. P.; Naveen Kumar C.; Pawanraj S. P.

Volume/Issue : Volume 11 - 2026, Issue 8 - August

Google Scholar : https://tinyurl.com/ykape7t7

DOI : https://doi.org/10.38124/ijisrt/26aug1211

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : Accurate real estate and land valuation is a challenging task due to the influence of multiple spatial, environmental, physical, infrastructural, and socioeconomic factors. Conventional property valuation methods often depend on manual inspection, historical transaction data, and expert judgment, which can be time-consuming, subjective, and difficult to scale. This paper proposes Estate Vision AI, an intelligent, video-first real estate and land analysis framework that integrates computer vision, machine learning, geospatial intelligence, and environmental feature analysis to provide automated property assessment and valuation. The proposed system accepts on-ground property or land videos as a primary input and extracts visual and contextual features related to land characteristics, built structures, accessibility, surrounding infrastructure, drainage conditions, vegetation, and other observable property attributes. These features are combined with location-based and market-related information to estimate property price, land value, and development potential. The framework additionally provides analysis of soil-related visual characteristics, water-flow and drainage indicators, flood-risk factors, transportation accessibility, nearby facilities, and utility availability. An automatic valuation mode enables prediction without requiring the user to provide a reference price, while a comparison mode supports comparative analysis between multiple properties. The system also incorporates location-based video retrieval to enhance contextual understanding of properties and surrounding areas. By combining visual, spatial, environmental, and market features within a unified framework, Estate Vision AI aims to reduce dependence on manual valuation, improve scalability and consistency, and provide an explainable decision-support tool for property buyers, sellers, investors, valuers, and planning authorities. The proposed framework demonstrates the potential of multimodal artificial intelligence for developing more efficient, data-driven, and geographically aware real estate valuation systems.

Keywords : Real Estate Valuation, Computer Vision, Machine Learning, Geospatial Intelligence, Property Price Prediction.

References :

1. R. B. Abidoye and A. P. C. Chan, “Artificial neural network in property valuation: Application framework and research trend,” Property Management, vol. 35, no. 5, pp. 554–571, 2017, doi: 10.1108/PM-06-2016-0027.
2. P. Jafary, D. Shojaei, A. Rajabifard, and T. Ngo, “Automated land valuation models: A comparative study of four machine learning and deep learning methods based on a comprehensive range of influential factors,” Cities, vol. 151, Art. no. 105115, 2024, doi: 10.1016/j.cities.2024.105115.
3. H. Lee, H. Han, C. Pettit, Q. Gao, and V. Shi, “Machine learning approach to residential valuation: A convolutional neural network model for geographic variation,” The Annals of Regional Science, vol. 72, pp. 579–599, 2024, doi: 10.1007/s00168-023-01212-7.
4. M. Koeva, O. Gasuku, M. Lengoiboni, K. Asiama, R. M. Bennett, J. Potel, and J. Zevenbergen, “Remote sensing for property valuation: A data source comparison in support of fair land taxation in Rwanda,” Remote Sensing, vol. 13, no. 18, Art. no. 3563, 2021, doi: 10.3390/rs13183563.
5. A. J. Bency, S. Rallapalli, R. K. Ganti, M. Srivatsa, and B. S. Manjunath, “Beyond spatial auto-regressive models: Predicting housing prices with satellite imagery,” in Proc. IEEE Winter Conference on Applications of Computer Vision (WACV), 2017, pp. 320–329, doi: 10.1109/WACV.2017.42.
6. P. Jafary, D. Shojaei, A. Rajabifard, and T. Ngo, “Automating property valuation at the macro scale of suburban level: A multi-step method based on spatial imputation techniques, machine learning and deep learning,” Habitat International, vol. 148, Art. no. 103075, 2024, doi: 10.1016/j.habitatint.2024.103075.
7. R. B. Abidoye and A. P. C. Chan, “Improving property valuation accuracy: A comparison of hedonic pricing model and artificial neural network,” Pacific Rim Property Research Journal, 2018, doi: 10.1080/14445921.2018.1436306.
8. R. B. Abidoye and A. P. C. Chan, “Modelling property values in Nigeria using artificial neural network,” Journal of Property Research, 2017, doi: 10.1080/09599916.2017.1286366.
9. R. B. Abidoye and A. P. C. Chan, “Critical review of hedonic pricing model application in property price appraisal: A case of Nigeria,” International Journal of Sustainable Built Environment, 2017, doi: 10.1016/j.ijsbe.2017.02.007.
10. R. B. Abidoye and A. P. C. Chan, “Valuers’ receptiveness to the application of artificial intelligence in property valuation,” Pacific Rim Property Research Journal, 2017, doi: 10.1080/14445921.2017.1299453.
11. P. Jafary, D. Shojaei, A. Rajabifard, and T. Ngo, “Automated valuation models and machine learning approaches for property and land valuation,” related research on automated valuation and spatial data integration, 2024.
12. Y. Kang, X. Zhang, S. Yao, Y. Liu, and S. Huang, “GeoAI for real estate valuation: Integrating spatial and visual information for property price prediction,” related research direction in computer vision and geospatial analysis.