Dehui Kong

Title: Dehui Kong: Innovator in Image Recognition and 3D Reconstruction

Introduction

Dehui Kong is a prominent inventor based in Beijing, China. He has made significant contributions to the fields of image recognition and 3D reconstruction. With a total of 3 patents, his work is at the forefront of technological advancements in these areas.

Latest Patents

One of his latest patents is the "Dictionary learning method and means for zero-shot recognition." This innovative method establishes the alignment between visual space and semantic space at both the category layer and image level, enabling high-precision zero-shot image recognition. The dictionary learning method involves several steps, including training a cross-domain dictionary of a category layer, generating semantic attributes of an image, and completing recognition tasks for invisible category images.

Another notable patent is the "3D reconstruction method based on deep learning." This method utilizes a potential vector constrained in the input image to reconstruct the complete 3D shape of a target. It learns the mapping between parts and the complete 3D shape, achieving high-precision reconstruction through various steps, including transforming predicted voxel floating values into binary values.

Career Highlights

Dehui Kong is affiliated with the Beijing University of Technology, where he continues to advance his research and innovations. His work has garnered attention for its practical applications in various industries, particularly in enhancing image recognition technologies.

Collaborations

He has collaborated with notable colleagues, including Shaofan Wang and Lichun Wang, contributing to a dynamic research environment that fosters innovation and development.

Conclusion

Dehui Kong's contributions to image recognition and 3D reconstruction highlight his role as a leading inventor in these fields. His innovative patents reflect a commitment to advancing technology and improving practical applications in real-world scenarios.