Jiteng Mu

Title: Jiteng Mu: Innovator in Image Synthesis Technology

Introduction

Jiteng Mu is a prominent inventor based in La Jolla, California. He has made significant contributions to the field of image synthesis, particularly through his innovative patent that focuses on learning dense correspondences for images. His work is instrumental in advancing the capabilities of generative adversarial networks (GANs).

Latest Patents

Jiteng Mu holds a patent titled "Learning Dense Correspondences for Images." This patent describes systems and methods that disentangle structure and texture representations of GAN synthesized images. The invention involves learning a dense pixel-level correspondence map for each image during the synthesis process. A canonical coordinate frame is defined, allowing for the warping of a structure latent code for each generated image to align with this frame. This approach establishes correspondences in a shared coordinate space, enabling the propagation of semantic labels from reference images to synthesized images.

Career Highlights

Jiteng Mu is currently employed at Nvidia Corporation, a leading company in graphics processing technology. His work at Nvidia has allowed him to explore and develop cutting-edge technologies in image processing and synthesis. His innovative approach has positioned him as a key player in the field.

Collaborations

Jiteng Mu collaborates with notable colleagues such as Sifei Liu and Shalini De Mello. Their combined expertise contributes to the advancement of technologies in image synthesis and related fields.

Conclusion

Jiteng Mu's contributions to the field of image synthesis through his patent and work at Nvidia Corporation highlight his role as an influential inventor. His innovative methods for learning dense correspondences for images are paving the way for future advancements in technology.