Ken Miyamoto

# Overview I have been working in Mitsubishi Electric Corporation's information technology center and business innovation. My role is primarily to develop technologies associated with 1. machine learning/AI, 2. computer vision, and 3. point cloud processing. The applications are visual inspection, measurement systems, robots, simulators, maintenance for electric equipment, and XR. I have issued some papers at international academic conferences (e.g., IEEE). # Background 1. Machine learning/AI： -Data augmentation: Data augmentation with keyframes of SLAM (simultaneous localization and mapping) -Anomaly detection : Anomaly detection for moving camera (ex: robot) -Versatile AI development: Semantic/panoptic Segmentation, object detection, 6DoF pose estimation, localization 2. Computer Vision / 3. Point cloud processing： 　-Visual SLAM, visual localization, structure from motion -Stereo matching, stereo intrinsic and extrinsic calibration with higher accuracy than open source 　-Point cloud viewer 　-Fusion of visual SLAM and IMU, pre-integration *Activity -2015-2016: IEC TC100 TA17 -2015: Join IEC young professional program as a Japanese delegate -2018: Present "AR in a Large Area through Instance Recognition with Hybrid Sensors" in IEEE VR 2018 -2020: Present "User-Selected Object Data Augmentation for 6DOF CNN Localization" in IEEE Consumer Electronics 2020 -2022-now: Teach machine learning and AI as an online tutor -2023: Present "Towards Robot Anomaly Detection" in British Machine Vision Conference -2024: Join IEC/ISO JTC1 * Language -TOEIC: 905