MUMBAI, India, Aug. 31 -- Intellectual Property India has published a patent application (202621082706 A) filed by Shabeen Narang; and Arvind Kumar Narang on July 05, 2026, for System And Method For Variance-Modulated Decision-Level Fusion In Multimodal Psychometric Assessment Systems.
Inventors include Shabeen Narang; and Arvind Kumar Narang.
The application for the patent was published on August 28, 2026, under issue no. 35/2026.
Abstract: ABSTRACT SYSTEM AND METHOD FOR VARIANCE-MODULATED DECISION-LEVEL FUSION IN MULTIMODAL PSYCHOMETRIC ASSESSMENT SYSTEMS A multimodal signal processing system processes a video stream and an audio stream associated with a human subject to generate a psychometric profile. A video-driven pipeline maps the video stream to a first set of personality trait scores. An independent audio-driven pipeline maps the audio stream to a second set of personality trait scores. A variance-modulated decision-level fusion engine computes a real-time kinematic variance signal from the video stream and a real-time acoustic variance signal from the audio stream. The fusion engine calculates normalized modality contribution weights based on an inverse relationship to the variance signals, blending the first and second sets of scores into final fused personality trait scores. A downstream perspective optimization layer selects specific scores based on a stored optimal perspective mapping to assemble a finalized profile. FIG. 1
Disclaimer: Curated by HT Syndication.
