Company Filing History:
Years Active: 2025
Title: Mohsen Zand: Innovator in Pose Estimation Technologies
Introduction
Mohsen Zand is a prominent inventor based in Kingston, CA. He has made significant contributions to the field of pose estimation through his innovative patent. His work focuses on enhancing object pose determination using advanced machine learning techniques.
Latest Patents
Mohsen Zand holds a patent titled "Systems and methods for pose estimation via radial voting based keypoint localization." This patent describes systems and methods that facilitate object pose determination through keypoint detection. The invention employs a machine learning algorithm to enable voting-based estimation of the locations of at least three keypoints. The algorithm is trained using reference intensity-depth images of an object to determine the radial distance between a keypoint and a 3D scene location associated with each pixel. During inference, the algorithm processes an intensity-depth image to generate radial distance estimates for each pixel. These estimates are used to increment an accumulator space, creating a sphere for each pixel centered on the 3D scene location. A keypoint location is determined by identifying a peak in the accumulator space, allowing for multiple keypoints to enable accurate pose determination. The methods described are adaptable to various imaging modalities, including non-depth images and point cloud datasets.
Career Highlights
Mohsen Zand is associated with Bluewrist Inc., where he continues to develop and refine his innovative technologies. His work has positioned him as a key figure in the field of pose estimation.
Collaborations
He has collaborated with notable professionals in the industry, including Michael Greenspan and Seyed Ali Etemad. These collaborations have further enhanced the impact of his work in the field.
Conclusion
Mohsen Zand's contributions to pose estimation technologies demonstrate his innovative spirit and commitment to advancing the field. His patent reflects a significant step forward in object pose determination, showcasing the potential of machine learning in this area.