Kun Duan

Title: Kun Duan: Innovator in Object Classification Technology

Introduction

Kun Duan is a notable inventor based in Bloomington, IN (US). He has made significant contributions to the field of image classification through his innovative patent. His work focuses on enhancing the accuracy and efficiency of object classification in digital images.

Latest Patents

Kun Duan holds a patent titled "Object classification with constrained multiple instance support vector machine." This patent provides methods and systems for classifying digital images of objects. Specifically, it describes a process where an object classifier is trained using a constrained MI-SVM (multiple instance-support vector machine) approach. This method involves sampling training images of objects to generate a collection of image regions associated with specific object types and viewpoints. The classifier is then trained to determine an appropriate mid-level representation of the training image that is discriminative.

Career Highlights

Kun Duan is currently employed at Xerox Corporation, where he continues to develop innovative solutions in image processing and classification. His work at Xerox has allowed him to apply his expertise in practical applications, contributing to advancements in technology.

Collaborations

Kun Duan collaborates with Luca Marchesotti, who is also involved in the field of image classification. Their partnership enhances the research and development efforts at Xerox Corporation.

Conclusion

Kun Duan's contributions to object classification technology demonstrate his commitment to innovation and excellence in the field. His patent reflects a significant advancement in the way digital images are classified, showcasing his role as a leading inventor in this area.