This inventor holds 2 USPTO granted patents. Top assignee: Hangzhou Dianzi University. Active years: 2024.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile
Company Filing History:
Years Active: 2024
Title: Innovations of Inventor Ming Jiang in Pedestrian Re-identification
Introduction
Ming Jiang, an inventive mind located in Hangzhou, China, has made significant contributions to the field of pedestrian re-identification. With a total of two patents under his name, his work focuses on enhancing the reliability and accuracy of pedestrian recognition methods using advanced technology.
Latest Patents
Ming Jiang's latest patents illustrate his innovative approach to solving complex problems related to pedestrian recognition. The first patent is titled "Transformer-based multi-scale pedestrian re-identification method." This invention introduces a pedestrian re-identification network utilizing multi-scale feature extraction and Transformers. The design comprises a multi-scale feature cascade module aimed at mining detailed features of pedestrians across various depths and scales, enabling stronger feature representation. Additionally, a Transformer-based feature extraction is incorporated to analyze pedestrian features on a global scale. The aggregated features output by the Transformer substantiate a superior representation of pedestrian characteristics, thus boosting the discrimination capabilities of the model. Tests demonstrate that this method exhibits enhanced robustness and adaptability, significantly improving the model’s generalization ability.
The second patent, "Occluded pedestrian re-identification method based on pose estimation and background suppression," addresses the challenges of recognizing occluded pedestrians. This method involves steps to extract both global and local features of occluded individuals, culminating in the creation of a local feature map. By generating a heat map of pedestrian image key-points and leveraging local features, the method enhances key-point features using a convolutional approach. An adjacency matrix of key-points further supports the extraction of final features, relying on both local and global data inputs to refine pedestrian recognition.
Career Highlights
Ming Jiang is affiliated with Hangzhou Dianzi University, where he continues to expand his research and innovative contributions. His roles in the university promote a strong intersection of academic rigor and practical application in technological advancements.
Collaborations
Ming Jiang collaborates with esteemed colleagues, including Lingjie He and Min Zhang, who are essential in diversifying the research and application of pedestrian identification technologies. These collaborations enhance the development processes, leading to more refined and effective solutions in the field.
Conclusion
Through his inventive patents and dedicated research at Hangzhou Dianzi University, Ming Jiang exemplifies the spirit of innovation. His work in pedestrian re-identification is paving the way for advancements in surveillance, security, and smart city applications, showcasing the vital role of inventors in shaping our interaction with technology.
Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile