MUMBAI, India, June 26 -- Intellectual Property India has published a patent application (202441098105 A) filed by Natoe, Inc on December 12, 2024, for System For Performance Evaluation And Improvement Of An Ai Model For Classifying Images In Real-Time.
Inventors include Rakesh Deshmukh; Gulshan Kumar; Pragya Mathur; and Herukh Shahani.
The application for the patent was published on June 19, 2026, under issue no. 25/2026.
Abstract: A system for evaluating and improving the performance of an artificial intelligence (AI) model (116) used for classifying images through a dual-pathway feedback mechanism is provided. The system includes a model performance evaluation and feedback server (106) having at least one processor (112) and a memory (114), wherein the processor (112) executes the AI model (116) to generate an AI-generated report containing preliminary findings from images received from one or more image capturing devices (102A-N). The server (106) receives expert-corrected reports and extracts structured correction data including the nature and severity of each correction. The structured correction data is correlated with a spatial attention map generated by the AI model (116) to identify a failure mode, including perceptual failures and interpretive failures. A performance report is generated based on the failure mode and severity level. The system performs a dual-pathway update, wherein perceptual failures trigger batch retraining of feature extraction layers, and interpretive failures result in logic constraint vectors stored in an inference guardrail database (118), enabling real-time inference correction without immediate weight modification. FIG. 1
Disclaimer: Curated by HT Syndication.
