Kanagawa, Japan

Hiroki Kawaski

This inventor holds 1 USPTO granted patent and 1 published patent application. Top assignee: International Business Machines Corporation. Active years: 2023.

IDiyas Innovation Intelligence. (2026). Inventor Profile: Hiroki Kawaski. Retrieved from https://idiyas.com/inventor/hiroki-kawaski

Data source: USPTO (United States Patent and Trademark Office) public patent records. Weekly synchronization. How IDiyas builds this profile

Calculating rank
USPTO Granted Patents = 1 

% Patents Active = 100.0

Average Co-Inventor Count = 2.0

ph-index = 1


Company Filing History:


Years Active: 2023

Loading Chart...
1 patent (USPTO):Explore Patents

Title: Hiroki Kawaski: Video Dataset Augmentation for Machine Learning

Introduction

Hiroki Kawaski is an inventor associated with Kanagawa, Japan, and International Business Machines Corporation. His listed patent work focuses on preparing video data for machine-learning training.

Latest Patents

Kawaski is credited with 1 patent: “Augmentation of Video Datasets for Machine Learning Training.” The invention describes dividing video frames into subframes and tallying how many pixels change over time in each subframe. The subframe with the greatest cumulative pixel change is selected as a cropping candidate. Cropped copies of the video can then be generated to include that candidate while excluding some surrounding portions, with transformed crop boundaries enabling additional variations.

Career Highlights

The available information identifies Kawaski’s work with International Business Machines Corporation and highlights his contribution to video-data augmentation for machine-learning algorithms. The described approach aims to create varied training datasets from existing video.

Collaborations

Shingo Nagai is identified as Kawaski’s coworker. The supplied information does not specify their particular projects or collaboration.

Conclusion

Hiroki Kawaski’s listed patent presents a method for generating cropped video variations based on regions with substantial pixel changes, supporting the preparation of machine-learning training data.

Profile summary based on public USPTO records.
Data Sources: USPTO Patent Grant XML, Patent Center, EPO & CIPO • Normalized by IDiyas Innovation Graph. Methodology & provenance architecturePlease report any incorrect information to [email protected]
Loading…