Peng Wang

Title: Innovations of Peng Wang in Speech and Depth Estimation Technologies

Introduction

Peng Wang is an accomplished inventor based in Sunnyvale, California, with a notable portfolio of six patents. His work primarily focuses on advanced technologies that enhance the interaction between humans and machines, particularly in the fields of speech processing and depth estimation.

Latest Patents

One of his latest patents is titled "Personalized speech-to-video with three-dimensional (3D) skeleton regularization and expressive body poses." This invention presents novel methods for converting speech audio or text into a photo-realistic speaking video of a person. The technology utilizes a recurrent neural network to generate 3D skeleton movements from audio sequences, while a conditional generative adversarial network synthesizes the output video. The incorporation of an articulated 3D human skeleton and a learned dictionary of personal speech iconic gestures ensures that the generated movements are both realistic and expressive.

Another significant patent is focused on "Systems and methods for depth estimation via affinity learned with convolutional spatial propagation networks." This invention aims to improve the speed and quality of real-time per-pixel depth estimation from a single image. By employing an end-to-end Convolutional Spatial Propagation Network (CSPN), the technology refines depth outputs and converts sparse depth samples into dense depth maps. This method is particularly beneficial for robotics and autonomous driving applications, where accurate depth measurements are crucial.

Career Highlights

Peng Wang has worked with prominent companies such as Baidu USA LLC and Baidu Times Technology (Beijing) Co. Ltd. His experience in these organizations has contributed significantly to his expertise in developing innovative technologies.

Collaborations

Throughout his career, Peng has collaborated with talented individuals, including Ruigang Yang and Xinjing Cheng. These partnerships have fostered a creative environment that has led to groundbreaking advancements in his field.

Conclusion

Peng Wang's contributions to speech processing and depth estimation technologies exemplify his innovative spirit and dedication to advancing human-computer interaction. His patents reflect a commitment to creating practical solutions that enhance the way we communicate and perceive our environment.