Hong Kong, China

Jiajun Shen

This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignees: Tcl Technology Group Corporation, University of Hong Kong Versitech Limited. Active years: 2026.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Jiajun Shen. Retrieved from https://idiyas.com/inventor/jiajun-shen-ziqq5xgr

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

USPTO Granted Patents = 1 

Average Co-Inventor Count = 1.0

ph-index = 1


Company Filing History:


Years Active: 2026

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Jiajun Shen and Video Temporal Action Proposal Generation.

Introduction

Jiajun Shen is an inventor associated with Hong Kong, China. His patent work includes a method and system for generating temporal action proposals in video.

Latest Patents

Shen is associated with 1 patent, titled “Method and System for Video Temporal Action Proposal Generation.” The invention processes features extracted from an input video through an encoder to produce representations containing global context. It then uses pre-trained proposal segments to extract corresponding features, which a decoder combines with proposal features to generate segment representations. A prediction module uses these representations to produce temporal action proposals.

The approach is designed to capture video-wide context and strengthen the representational capacity of the encoded features. Its learnable proposal segments also help accelerate training convergence and reduce computational burden.

Career Highlights

Shen’s listed company affiliations include the University of Hong Kong Versitech Limited and TCL Technology Group Corporation.

Collaborations

His listed coworkers include Ping Luo and Jiannan Wu.

Conclusion

Jiajun Shen’s patent focuses on improving video action proposal generation through global-context encoding and learnable proposal segments, with the aim of making the process more effective and computationally efficient.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…