Jiabin Zhu

Title: Jiabin Zhu: Inventor in Video Retrieval and Human-Machine Collaboration

Introduction

Jiabin Zhu is an inventor associated with Hangzhou Dianzi University in Hangzhou, China. Her work includes a patented approach to identifying complex targets in large video collections through collaboration between people and machine-vision systems.

Latest Patents

Zhu is associated with 1 patent: “Method for Complex Target Identification from Mass Videos Based on Human-Machine Collaboration.” The invention addresses the limitations of machine-vision systems when environments change or targets are obscured or camouflaged. It first uses a retrieval model to screen a video library based on coarse-grained target features. The resulting candidates are then presented to a user through a brain-eye synergistic rapid serial visual presentation (RSVP) process. After the user identifies the target, its specific features are supplied to the retrieval model to support faster location and tracking across the video collection.

Career Highlights

Zhu’s patent focuses on combining automated video screening with human visual judgment. This approach is intended to reduce the time and labor required to find targets when machine-vision systems have difficulty generalizing to changing or challenging conditions.

Collaborations

Zhu is associated with Hangzhou Dianzi University. Her listed coworkers include Xuanyu Jin and Li Zhu.

Conclusion

Jiabin Zhu’s patent presents a human-machine collaboration method for improving complex target retrieval in large video collections.