This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: Inception AI IP Ltd. Active years: 2026.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2026
Title: Mohammad Derakhshani: Image Categorization with Visual Language Models.
Introduction
Mohammad Derakhshani is an inventor based in Amsterdam, Netherlands. His work includes a patented method for using a visual language model to categorize images.
Latest Patents
Derakhshani is associated with 1 patent, titled “Method and System for Image Categorization Using a Visual Language Model.” The invention groups images into clusters representing different object categories, assigns names to those clusters, and generates captions for them. The resulting image-caption pairs provide context to a visual language model, which receives a query asking it to categorize a new image from a represented class. The model then generates an open-ended categorization in response.
Career Highlights
Derakhshani is associated with Inception AI IP Ltd. His patent describes an approach that combines image clustering, descriptive captions, and contextual prompting to support image categorization.
Collaborations
His listed coworkers include Ivona Najdenkoska and Yuki Asano.
Conclusion
Mohammad Derakhshani’s patent presents a context-driven method for image categorization that uses image-caption pairs to guide a visual language model’s response.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile