This inventor holds 1 USPTO granted patent and 2 published patent applications. Top assignees: Carnegie Mellon University, Toyota Jidosha Kabushiki Kaisha, Toyota Research Institute, Inc.. Active years: 2026.
Company Filing History:
1 out of 1,181 
1 out of 38,723 
1 out of 865 
Years Active: 2026
Title: Junyu Nan - Innovator in Point Cloud Forecasting
Introduction
Junyu Nan is a prominent inventor based in Pittsburgh, PA (US). He has made significant contributions to the field of point cloud forecasting, showcasing his expertise through innovative methodologies. His work is characterized by a focus on advanced machine learning techniques and their applications in various industries.
Latest Patents
Junyu Nan holds a patent for a "Method for diverse sequential point cloud forecasting." This patent describes a sophisticated approach that includes training a vector-quantized conditional variational autoencoder (VQ-CVAE) framework. The method aims to map an output to the closest vector in a discrete latent space to obtain a future latent space. It also involves outputting a categorical distribution of the probability of V vectors in response to previously sampled latent spaces and past point cloud sequences. Furthermore, the method includes sampling an inferred future latent space and predicting future point cloud sequences while incorporating denoising techniques.
Career Highlights
Throughout his career, Junyu Nan has worked with notable organizations, including the Toyota Research Institute and Toyota Motor Corporation. His experience in these companies has allowed him to apply his innovative ideas in real-world scenarios, contributing to advancements in automotive technology and machine learning.
Collaborations
Junyu has collaborated with talented individuals such as Xinshuo Weng and Jean Mercat. These partnerships have fostered a creative environment that encourages the exchange of ideas and the development of cutting-edge technologies.
Conclusion
Junyu Nan's contributions to point cloud forecasting and his innovative methodologies highlight his role as a leading inventor in the field. His work continues to influence advancements in technology and machine learning applications.