Jing Li

Title: Jing Li: Innovator in Motion Prediction Technology

Introduction

Jing Li is a notable inventor based in Richardson, TX (US), recognized for his contributions to the field of motion prediction technology. With a focus on enhancing image processing through innovative architectures, he has made significant strides in the application of convolutional neural networks (CNNs).

Latest Patents

Jing Li holds 1 patent for his invention titled "Cascaded architecture for disparity and motion prediction with block matching and convolutional neural network (CNN)." This patent describes a CNN that operates on the disparity or motion outputs of a block matching hardware module, such as a DMPAC module. The technology aims to produce refined disparity or motion streams, improving operations in images with ambiguous regions. By leveraging the processing capabilities of the block matching hardware module, the CNN can remain small and operate in real time, contrasting with traditional CNNs that handle all processing tasks. Notably, the CNN's operation is triggered only when the block hardware module output confidence level falls below a predetermined threshold, allowing for various configurations while maintaining real-time performance on conventional platforms.

Career Highlights

Jing Li is currently employed at Texas Instruments Corporation, where he continues to develop innovative solutions in the realm of image processing and motion prediction. His work has positioned him as a key player in advancing technology that enhances the efficiency and accuracy of image analysis.

Collaborations

Jing collaborates with talented colleagues, including Tarek Aziz Lahlou and Do-Kyoung Kwon, who contribute to the innovative environment at Texas Instruments Corporation.

Conclusion

Jing Li's work exemplifies the intersection of technology and innovation, particularly in the field of motion prediction. His contributions are paving the way for advancements that improve image processing capabilities in various applications.