Ming Kai Hsu

Title: Innovations of Ming Kai Hsu

Introduction

Ming Kai Hsu is a prominent inventor based in Fremont, California. He holds a total of six patents, showcasing his contributions to the field of technology and innovation. His work primarily focuses on advancements in video processing and neural networks.

Latest Patents

One of his latest patents is titled "Class-specific neural network for video compressed sensing." This invention provides methods for training and testing a class-specific neural network that includes a Gaussian-mixture model (GMM). The GMM classifies video frame blocks into multiple clusters and assigns them accordingly. The encoders then generate compressed-sensed frame block vectors based on these clusters. Another significant patent is "Generative adversarial network for video compression." This method involves a decoding terminal that receives coded video frames processed using generative adversarial networks (GANs). The terminal decodes these frames based on network parameters, implementing various video coding functions.

Career Highlights

Ming Kai Hsu has worked with notable companies such as Cadence Design Systems, Inc. and Kwai Inc. His experience in these organizations has contributed to his expertise in video processing technologies and neural networks.

Collaborations

Some of his coworkers include Ying Liu and Nam Ling, who have collaborated with him on various projects throughout his career.

Conclusion

Ming Kai Hsu's innovative work in video processing and neural networks has led to significant advancements in the field. His contributions are reflected in his patents and collaborations, marking him as a notable figure in technology.