Company Filing History:
Years Active: 2023-2024
Title: The Innovations of Deqing Sun
Introduction
Deqing Sun, an innovative inventor based in Cambridge, MA, has made significant contributions to the field of image processing and computer vision. With two patents to his name, Sun’s work showcases his ability to blend advanced technology with practical applications, paving the way for new advancements in visual data analysis.
Latest Patents
Deqing Sun's latest patents reflect his knowledge and creativity in developing systems and methods for image correspondences. One of his notable patents is titled "Systems and methods for training models to predict dense correspondences in images using geodesic distances." This invention leverages synthetic training data created from 3D computer models, utilizing geodesic distances derived from these models to enhance model training.
Another remarkable patent, "Learnable cost volume for determining pixel correspondence," involves a sophisticated method for establishing correspondences between images. This patent details a process whereby feature vectors are transformed and combined to generate a cost volume, ultimately facilitating pixel correspondence determination in a seamless manner.
Career Highlights
Sun is currently employed at Google Inc., where he continues to apply his expertise in developing algorithms and systems that push the boundaries of what is possible in image analysis. His role at such a highly regarded company underscores his importance in the field and highlights his contributions to ongoing technological advancements.
Collaborations
Throughout his career, Deqing Sun has collaborated with other talented researchers, notably Taihong Xiao and Ming-Hsuan Yang. These partnerships demonstrate the collaborative spirit in the research community, contributing to the progression of innovative solutions in image processing.
Conclusion
Deqing Sun's innovative work in image processing has the potential to transform how images are analyzed and understood. With a focus on predictive modeling and cost volume analysis, his patents represent a significant step towards improving the accuracy and efficiency of image correspondence detection. As he continues to work at Google Inc., the impact of his inventions is likely to resonate within the tech industry and beyond.