General Information

Full Name Isaac Kasahara


  • 2022
    Master of Science in Robotics
    University of Minnesota
    • Implemented traditional and deep learning computer vision methods such as, camera pose estimation, multi-view camera calibration, bundle adjustment, stereo vision, key-point detection, segmentation, and more.
    • Acquired hands-on experience in applying common robotics algorithms such as SLAM, wall detection, visual servoing, forward kinematics, inverse kinematics, quaternions, etc.
  • 2019
    Bachelor of Science in Computer Science
    University of Minnesota
    • Acquired a solid understanding of essential data structures like arrays, linked lists, stacks, queues, trees, and graphs.
    • Manipulated and searched the aformentioned datastructures with algorithms such as sorting algorithms, search algorithms, greedy algorithms, etc.


  • 2023-Present
    Research Engineer
    Samsung AI Center NYC
    • Worked on both integrating computer vision algorithms onto real robots/devices as well as developing novel contributions to the field of computer vision.
    • Worked on projects such as
      • Fine-grained Control for Pose Conditioned Image Generation
      • 6D Object Pose Estimation for Robotic Grasping
      • Generalizable 3D Scene Reconstruction
      • Tracking/Following Humans with Mobile Robots
      • LiDAR from Mobile Robot to Floorplan Generation
  • 2022-2023
    Research Intern
    Samsung AI Center NYC
    • Worked on implementing real-time practical computer vision algorithms for a physical robot arm grasping challenge.
      • Focused in estimating 3D locations of glass objects for grasping.
      • Retraining a 6D pose/shape estimation network on in-house data.
    • Created a realistic synthetic dataset using PyBullet and PyRender distributed across multiple computers/gpus using Ray.
  • 2021-2022
    Research Assistant
    University of Minnesota
    • Worked under Professor Hyun Soo Park in collaboration with Toyota Research Institute to better understand driver attention.
      • Collected a large multi-camera dataset of real drivers on real roads.
      • Learned about camera calibration as well as synchronization.
      • Implemented a self-supervised learning approach to estimate both driver gaze and road scene saliency.
      • Published paper in European Conference for Computer Vision 2022 and publicly released dataset.
  • 2019
    Software Engineer
    Cirrus Aircraft
    • Developed strong coding/communication skills while learning about software requirements for aircraft avionics components.
  • 2018
    Software Intern
    Cirrus Aircraft
    • Worked on developing algorithms for experimental sensors on the SF50 Jet.

Honors and Awards

  • 2022
    • Selected to present in front of 200 people for Oral Presentation ECCV 2022
  • 2019
    • Dean's List at University of Minnesota


  • Computer Vision
    • Tools - Python, MATLAB, OpenCV, PyTorch, SciKit Learn, Open3D
    • Skills - Segmentation, 6D Pose Estimation, Stereo Vision, Tracking, Camera Calibration
  • Machine Learning
    • Tools - Python, Pytorch, Ubuntu, Linux, Github, Weights&Biases, Cuda, Pandas, Numpy
    • Skills - Training/Evaluation of Deep Networks, CNNs, Self-Supervised Learning, LSTMs
  • Robotics
    • Tools - ROS, PyBullet, PyRender, Ubuntu, Linux
    • Skills - SLAM, Camera Calibration, Grasp Estimation

Other Interests

  • Tennis, Pickleball, Bouldering, Movies, and Music.