Hyeok Min Choi

Title: Hyeok Min Choi and Deep Learning Dataset Augmentation.

Introduction

Hyeok Min Choi is an inventor associated with Seongnam, South Korea, and the Korea Electronics Technology Institute. Choi’s patent work addresses training data augmentation for deep learning systems that analyze image regions identified by a detector.

Latest Patents

Choi is associated with 1 patent: “Training Dataset Augmentation Method and System for Training Deep Learning Network.” The invention configures a training dataset using image data containing a detected area and its label, then creates additional training data by transforming that area. Randomly transforming detected regions is intended to help a deep learning model resist performance degradation caused by variations in detector performance.

Career Highlights

Choi’s documented patent contribution focuses on improving the robustness of deep learning models through dataset augmentation. The available information identifies the Korea Electronics Technology Institute as Choi’s organization.

Collaborations

The listed coworkers are Jin Woo Jeong and Sung Jei Kim. The provided information does not specify their roles in the patent or their particular contributions.

Conclusion

Hyeok Min Choi’s patent describes a practical approach to strengthening deep learning models by expanding training datasets with transformed image regions.