This inventor holds 4 USPTO granted patents and 3 published patent applications. Top assignees: Toyota Jidosha Kabushiki Kaisha, Toyota Research Institute, Inc., Carnegie Mellon University. Active years: 2026.
Company Filing History:







Years Active: 2026
Title: Pavel Tokmakov: Innovator in Video Understanding and Object Interaction
Introduction
Pavel Tokmakov is a prominent inventor based in Santa Monica, CA (US). He has made significant contributions to the fields of video understanding and object interaction, holding a total of 4 patents. His innovative methods are paving the way for advancements in artificial intelligence and machine learning.
Latest Patents
Tokmakov's latest patents include a method for self-supervised compositional feature representation for video understanding. This method involves processing a sequence of video frames through a feature map network to generate multiple feature maps. It utilizes an attention model to bind these feature maps to a fixed set of slot variables based on a motion segmentation signal. The combined feature map is then decoded to reconstruct a sequence of video frames, allowing for the identification of objects within the video.
Another notable patent focuses on estimating the trajectory of an object in an environment. This method identifies uncertainty regions using a keypoint classifier and plans interactions with the object based on these regions. The planned interaction occurs in a different area of the environment, ensuring effective engagement with the object.
Career Highlights
Throughout his career, Tokmakov has worked with esteemed organizations such as the Toyota Research Institute and Toyota Motor Corporation. His experience in these companies has allowed him to refine his skills and contribute to groundbreaking research in his field.
Collaborations
Tokmakov has collaborated with notable professionals, including Shuran Song and Simon Stent. These partnerships have further enhanced his work and led to innovative solutions in technology.
Conclusion
Pavel Tokmakov is a distinguished inventor whose work in video understanding and object interaction is shaping the future of artificial intelligence. His patents reflect a deep understanding of complex systems and a commitment to innovation.