MUMBAI, India, June 30 -- Intellectual Property India has published a patent application (202641076358 A) filed by Sr University on June 19, 2026, for A Semantic Causality And Knowledge Fusion System For Explainable Vision-Language Image Captioning.
Inventors include Eranti Divya; and Dr. Johnson Kolluri.
The application for the patent was published on June 26, 2026, under issue no. 26/2026.
Abstract: ABSTRACT Disclosed herein is a semantic causality and knowledge fusion system (100) for explainable vision–language image captioning, the system (100) comprises an image acquisition module (102) configured to receive input images. A visual feature extraction module (104) configured to detect and extract visual features. A semantic knowledge acquisition module (106) configured to obtain semantic information. A multimodal representation generation module (108) configured to transform the extracted visual features and the obtained semantic information. A semantic causality module (110) configured to identify causal dependencies, relational interactions, event associations, contextual influences, and logical relationships. A knowledge fusion module (112) configured to combine visual information, semantic information, contextual knowledge, and causal relationship information. A caption generation module (114) configured to generate image captions. An explainability module (116) configured to generate reasoning outputs. An output module (118) configured to provide the image captions together with the reasoning outputs.
Disclaimer: Curated by HT Syndication.
