This inventor holds 1 USPTO granted patent. Top assignee: Sap Se. Active years: 2023.
Company Filing History:
Years Active: 2023
Title: Hasnain Raza: Innovator in Machine Learning and Image Segmentation
Introduction
Hasnain Raza is a prominent inventor based in Berlin, Germany. He has made significant contributions to the field of machine learning, particularly in image segmentation. His innovative approach has led to the development of a unique patent that enhances the capabilities of machine learning models.
Latest Patents
Hasnain Raza holds a patent for "Weakly supervised one-shot image segmentation." This invention involves a machine learning model trained on a diverse set of training images that depict various base objects. Each training image is associated with a ground-truth segmentation that corresponds to one or more pixel-wise labels. The model learns base class prototypes related to segmentations of similar base objects. It is further trained using a support image that depicts a novel object, allowing the model to learn a novel class prototype based on similarities identified with the base class prototype. This trained model can then be applied to segment a query image, providing a robust solution for image analysis.
Career Highlights
Hasnain Raza is currently employed at SAP SE, where he continues to push the boundaries of technology and innovation. His work at SAP SE allows him to collaborate with other talented professionals in the field, contributing to advancements in software and machine learning applications.
Collaborations
Some of Hasnain's notable coworkers include Moin Nabi and Tassilo Klein. Their collaborative efforts in the field of machine learning and image processing have further enhanced the impact of their work.
Conclusion
Hasnain Raza is a key figure in the realm of machine learning and image segmentation. His innovative patent and contributions to SAP SE highlight his commitment to advancing technology. His work continues to inspire and pave the way for future innovations in the field.
