Company Filing History:
Years Active: 2022
Title: The Innovative Contributions of Daniel R. Mendat
Introduction
Daniel R. Mendat is a notable inventor based in Baltimore, MD (US). He has made significant contributions to the field of action recognition through his innovative patent. His work combines advanced technologies to enhance the understanding of human actions.
Latest Patents
Daniel R. Mendat holds a patent titled "Systems and method for action recognition using micro-doppler signatures and recurrent neural networks." This patent presents systems and methods for action recognition developed using a multimodal dataset that incorporates both visual data and active acoustic data. The dataset includes twenty-one actions and focuses on examples of orientational symmetry that a single active ultrasound sensor should have the most difficulty discriminating. The combined results from three independent ultrasound sensors are encouraging and provide a foundation to explore the use of data from multiple viewpoints to resolve the orientational ambiguity in action recognition. In various embodiments, recurrent neural networks using long short-term memory (LSTM) or hidden Markov models (HMMs) are disclosed for use in action recognition, particularly human action recognition, from micro-Doppler signatures. He has 1 patent to his name.
Career Highlights
Daniel R. Mendat is affiliated with Johns Hopkins University, where he continues to contribute to research and innovation in his field. His work is characterized by a strong emphasis on integrating different types of data to improve action recognition systems.
Collaborations
Daniel has collaborated with notable colleagues such as Andreas G. Andreou and Kayode Sanni, enhancing the depth and breadth of his research efforts.
Conclusion
Daniel R. Mendat's innovative work in action recognition showcases the potential of combining visual and acoustic data to advance technology in this area. His contributions are paving the way for future developments in understanding human actions through sophisticated systems.