Intellectual Property India Publishes Patent Application for 'System And Method For Categorization Of Indian Art Styles Using Deep Convolutional Neural Networks With Multi-Modal Similarity Analysis' Filed by Vellore Institute Of Technology

MUMBAI, India, Feb. 27 -- Intellectual Property India has published a patent application (202641018067 A) filed by Vellore Institute Of Technology, Vellore, Tamil Nadu, on Feb. 18, for 'system and method for categorization of indian art styles using deep convolutional neural networks with multi-modal similarity analysis.'

Inventor(s) include Dr. G. Anuradha; and Mr. G. Ramachandra Aditya.

The application for the patent was published on Feb. 27, under issue no. 09/2026.

According to the abstract released by the Intellectual Property India: "A system (4200) for categorizing painting styles and analyzing inter-style similarities includes a data input layer (4202) configured to receive painting images representing distinct painting styles, a preprocessing submodule (4206) configured to resize painting images and normalize pixel values, a feature extraction module (4208) with a convolutional neural network (4210) configured to extract deep feature vectors from each preprocessed painting image, a feature store (4212) configured to store the extracted feature vectors, a classification module (4214) with a classification head (4216) configured to classify painting images into distinct painting styles based on the extracted feature vectors, and a similarity analysis module (4220) with a similarity computation submodule (4222), a clustering submodule (4226), a dominant color analysis submodule (4228), and a shape detection submodule (4230), wherein the system (4200) achieves multi-modal similarity analysis by combining feature vectors with dominant colors and geometric shapes."

Disclaimer: Curated by HT Syndication.