R&D
Publications
Please see my Google Scholar profile which should maintain an up-to-date list of all my publications with full citation information.
Self-Supervised 3D Keypoint Learning for Ego-Motion Estimation
Jiexiong Tang, Rares Ambrus, Vitor Guizilini, Sudeep Pillai, Hanme Kim, Patric Jensfelt, Adrien Gaidon
4th Conference on Robot Learning (CoRL 2020), Cambridge MA, USA. 2020
Neural Outlier Rejection for Self-Supervised Keypoint Learning
Jiexiong Tang, Hanme Kim, Vitor Guizilini, Sudeep Pillai, Rares Ambrus
The International Conference on Learning Representations (ICLR), Addis Ababa, Ethiopia, 2020
Real-Time Visual SLAM with an Event Camera
Hanme Kim
PhD Thesis, Imperial College London, UK, 2017
Real-Time 3D Reconstruction and 6-DoF Tracking with an Event Camera
Hanme Kim, Stefan Leutenegger, and Andrew J. Davison
European Conference on Computer Vision (ECCV), Amsterdam, The Netherlands, 2016 (Best Paper Award) [Presentation] | [Poster]
Towards Visual SLAM with Event-based Cameras
Michael Milford, Hanme Kim, Stefan Leutenegger, and Andrew J. Davison
The Problem of Mobile Sensors: Setting future goals and indicators of progress for SLAM Workshop, Robotics Science and Systems (RSS), Rome, Italy, 2015
Place Recognition with Event-based Cameras and a Neural Implementation of SeqSLAM
Michael Milford, Hanme Kim, Michael Mangan, Stefan Leutenegger, Tom Stone, Barbara Webb, Andrew J. Davison
The “Innovative Sensing for Robotics: Focus on Neuromorphic Sensors” workshop,
IEEE International Conference on Robotics and Automation (ICRA),
Seattle, Washington, USA, 2015
Simultaneous Mosaicing and Tracking with an Event Camera
Hanme Kim, Ankur Handa, Ryad Benosman, Sio-Hoi Ieng, and Andrew J. Davison
British Machine Vision Conference (BMVC), Nottingham, UK, 2014 (Best Industry Paper) [Presentation]
Patents
Systems and Methods for Detecting and Matching Keypoints Between Different Views of a Scene
Jiexiong Tang, Rares A Ambrus, Vitor Guizilini, Sudeep Pillai, Hanme Kim
US Patent 11,531,892. Dec. 20, 2022
Systems and Methods for Training a Neural Keypoint Detection Network
Jiexiong Tang, Rares A Ambrus, Vitor Guizilini, Sudeep Pillai, Hanme Kim
US Patent 11,256,986. Feb. 22, 2022
Incremental Map Building Using Learnable Features and Descriptors
Jiexiong Tang, Rares Andrei Ambrus, Hanme Kim, Vitor Guizilini, Adrien David Gaidon, Xipeng Wang, Jeff Walls, Sudeep Pillai
UUS Patent App. 17/230,942. Oct. 14, 2021
Self-Supervised 3D Keypoint Learning for Monocular Visual Odometry
Jiexiong Tang, Rares A Ambrus, Vitor Guizilini, Sudeep Pillai, Hanme Kim, Adrien David Gaidon
US Patent App. 17/093,393. Aug. 5, 2021
Self-Supervised 3D Keypoint Learning for Ego-Motion Estimation
Jiexiong Tang, Rares A Ambrus, Vitor Guizilini, Sudeep Pillai, Hanme Kim, Adrien David Gaidon
US Patent App. 17/093,360. Aug. 5, 2021
Software
SceneLib2 [GitHub Repository]
Open-source C++ library for MonoSLAM originally designed and implemented by Prof. Andrew Davison and colleagues at the University of Oxford.
Video
Please see my YouTube channel which should maintain an up-to-date list of all demo videos.
ETAM 2D: Real-Time Event-Based Tracking and Mapping with an Event Camera
Event-Based Camera Calibration
Event-Based Camera vs Standard Camera
We recreated this animation inspired by the associated animation of Mueggler, Huber, and Scaramuzza, “Event-based, 6-DOF Pose Tracking for High-Speed Maneuvers”, IROS, 2014
Past Projects
Mostly industrial projects which I worked on in my previous companies.
- ANPR Engine for Embedded (NDI-RS, UK)
- ANPR Engine for Windows (NDI-RS, UK)
- Upgrade ANPR Engine (NDI-RS, UK)
- ANPR Engine using OpenCV (Personal, UK)
- Smart Laser Sensor (Meta Vision Systems, UK)
- Upgrade Digital Laser Scanner (Meta Vision Systems, UK)
- Image/Video Compression (Eyenix, South Korea)
- Custom ASIC HD/SD Scaler (Eyenix, South Korea)
- Active Tracking & Image Mosaicing (Eyenix, South Korea)
- Audio/Video Mixer (Inter-M, South Korea)
- Zoom Camera (Inter-M, South Korea)
- CCTV Cameras & Systems (Inter-M, South Korea)
- Segmentation & Motion Tracking (Myongji University, South Korea)
- Medical Image Segmentation (Myongji University, South Korea)
- Frame Grabber Application (Myongji University, South Korea)