Hang Su

Title: Innovations of Hang Su

Introduction

Hang Su is an accomplished inventor based in Hadley, MA (US). She has made significant contributions to the field of technology, particularly in image processing and 3D object analysis. With a total of 2 patents, her work has garnered attention in the industry.

Latest Patents

One of her latest patents is titled "View synthesis using attribute correspondences and geometric relationship constraints." This innovation allows for the synthesis of a target image corresponding to a novel view from two source images. The method utilizes pixel attribute correspondences and geometric relationships to define each pixel in the target image based on identified pixels in the source images. This approach ensures that a particular object in the target image can be visible in just one of the source images for successful synthesis.

Another notable patent is "Bilateral convolution layer network for processing point clouds." This method involves filtering a point cloud transformation of a 3D object to create a 3D lattice. The lattice is then processed through a series of bilateral convolution networks, each with a lower lattice feature scale than the previous one. The output from each network is concatenated to generate an intermediate 3D lattice, which is further filtered to predict features of the 3D object.

Career Highlights

Hang Su is currently employed at Nvidia Corporation, a leading company in graphics processing technology. Her work at Nvidia has allowed her to push the boundaries of what is possible in image synthesis and 3D processing.

Collaborations

Some of her notable coworkers include Varun Jampani and Deqing Sun, who have also contributed to advancements in the field of technology.

Conclusion

Hang Su's innovative patents and contributions to Nvidia Corporation highlight her role as a significant figure in the realm of technology. Her work continues to influence advancements in image processing and 3D object analysis.