MUMBAI, India, Sept. 22 -- Intellectual Property India has published a patent application (202621093401 A) filed by Rk University on July 31, 2026, for Machine Learning-Based System And Method For Automatic Yoga Pose Detection And Posture Correction.
Inventors include Amit Lathigara; Nirav Bhatt; Paresh Tanna; Homera Durani; and Chhaya Patel.
The application for the patent was published on September 18, 2026, under issue no. 38/2026.
Abstract: Abstract MACHINE LEARNING-BASED SYSTEM AND METHOD FOR AUTOMATIC YOGA POSE DETECTION AND POSTURE CORRECTION A system for automatic yoga pose detection and posture correction includes a Camera Interface, a Body Landmark Detector, a Pose Classifier, a Joint Deviation Analyzer, and a Correction Guidance Interface. The system receives camera frames, body landmarks, joint angles, pose-sequence state, and user calibration data. The acquired observations are checked for quality, source association, and timing before a model-ready representation is generated. The Pose Classifier produces an analytical result and a confidence indication. The Joint Deviation Analyzer applies a stored response criterion and determines whether to release an output, request reacquisition, or require review. The Correction Guidance Interface provides a detected pose and directional posture-correction guidance. The system particularly performs comparing pose-specific joint angles with stored tolerance ranges and generating joint-directed correction instructions. An event record stores input identifiers, a processing version, the analytical result, and the delivered response to maintain traceability between acquisition, processing, and output. Fig. 1 Dated July 30, 2026 / Mungalpara Jigneshbhai Chhaganbhai IN/PA-2640 Agent for the Applicant
Disclaimer: Curated by HT Syndication.
