Boqing Gong

Title: Boqing Gong: Innovator in Video Quality Assessment and Captioning

Introduction

Boqing Gong is a prominent inventor based in Bellevue, WA (US). He has made significant contributions to the field of video technology, particularly in video quality assessment and automatic video captioning. With a total of 5 patents, Gong's work is at the forefront of innovation in multimedia processing.

Latest Patents

Gong's latest patents include a highly efficient model for video quality assessment. This method involves generating reduced-quality streams from a video stream of a first spatial and temporal resolution. By sampling subsets of spatial-temporal points (STPs) from these streams, a machine learning model processes the STPs to identify quality scores. These scores help in determining quality-improving actions for the video stream. Another notable patent is for end-to-end video captioning using multi-task reinforcement learning. This method trains a captioning model to perform automatic video captioning by utilizing long short-term memory (LSTM) units and convolutional neural networks (CNNs). The model is initialized with cross-entropy loss and trained using reinforcement learning to generate accurate video captions.

Career Highlights

Throughout his career, Boqing Gong has worked with leading technology companies, including Tencent America LLC and Google Inc. His experience in these organizations has allowed him to refine his skills and contribute to groundbreaking projects in video technology.

Collaborations

Gong has collaborated with talented individuals such as Yilin Wang and Miao Yin. These partnerships have fostered innovation and creativity in his projects.

Conclusion

Boqing Gong is a distinguished inventor whose work in video quality assessment and captioning has made a significant impact in the field. His innovative patents and collaborations highlight his commitment to advancing technology in multimedia processing.