This inventor holds 1 USPTO granted patent and 2 published patent applications. Top assignee: Tianjin University. Active years: 2022.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2022
Title: Dan Song: Inventor in Visual Relationship Detection
Introduction
Dan Song is an inventor based in Tianjin, China, and is associated with Tianjin University. Her work includes a patent on visual relationship detection using region-aware learning mechanisms.
Latest Patents
Song is listed as an inventor on Visual Relationship Detection Method and System Based on Region-Aware Learning Mechanisms. The invention describes a method that builds graph structures from object relationships and their features, then uses visual attention and message propagation to infer predicates between object sets. The method combines information from neighboring nodes with visual features from the regions associated with entity objects. Song is credited with 1 patent.
Career Highlights
The available information identifies Song’s patent work in visual relationship detection and her association with Tianjin University. The patent focuses on combining graph-based representations with region-level visual features to support relationship inference.
Collaborations
Song’s listed coworkers include Ning Xu and Hongshuo Tian.
Conclusion
Dan Song’s patent reflects research into graph-based learning and visual features for detecting relationships between objects.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile