【Food Manipulation】Image Processing Engineer【Vision-based Tactile Sensor】

Machine Learning (inc. Deep Learning) | Computer Vision, Image Recognition | Manipulation | Embedded Software

Location	Koto-ku, Tokyo or Kyoto-shi, Kyoto-fu
Company	A startup whose core technology is vision-based tactile sensor that enables multi-modal perception at low cost and with high durability. Working on its application to pick-and-place automation in ready-made food production. Foundation: October 2021 Capital: 104 million yen
Job Description	This role processes images acquired from external cameras and sensors and converts them into the information necessary for motion generation. The role also covers a wide range of image processing, from low-level processing such as feature point tracking to image recognition such as object segmentation, as well as GPU implementation and embedded programming.
Requirement	Required skills: - Deep learning for computer vision (segmentation, classification, anomaly detection, etc.) - Traditional computer vision skills (non-deep learning) - OpenCV - PyTorch (or other deep learning framework) - C++ - Python Preferred skills: - ROS - Image processing on embedded systems - Robot learning with computer vision
English Level	Either English or Japanese at business level required
Japanese Level	Either English or Japanese at business level required
Employment Type	Permanent

Less than half of all openings are posted.
We may have prospective / non-public opportunities to introduce to registrants.
(Click here to register)