CV

Experiences

• May 2020 to March 2021 : Research Assistant supported by General Dynamics
  • Build a visual system to pixel-wise localize camera image of UAV with satellite map.
  • The system is able to simultaneously learn invariance features from multi-modality image pairs.
  • Several computer vision technics are seamlessly assembled together, including hierarchy Lucas-Kanade, Homography Mapping, Spatial Transform Network, etc.
• May 2019 to November 2019: Robotic Visual Perception Intern at iRobot
  • Build a label-free training system for monocular depth prediction network.
  • The system works stably for both indoor and outdoor cases by combining 3D reconstruction (colmap) with self-supervised learning.
• Other Semesters of my PhD: Teaching Assistant at Worcester Polytechnic Institute
  • Course co-designer of two graduate level courses: ECE579 and ECE545/CS545 which cover topics in machine learning and computer vision.
  • General teaching assistant workload (grade, office hour, lab manager, etc) for several undergraduate courses: ECE2049, ECE 2029 and ECE 2010.

Skills

• Programming skills
  • language: Python, C++, C, R, sql.
  • tools & frameworks: Pybind11, Tensorflow 1&2, PyTorch, ROS, Git, Markdone.
  • Learned at class: Fortran, Java, Scala, MongoDB, Hadoop, Spark, etc.
• Mathematical skills:
  • Foundatoins: Real & Complex Analysis, Linear & Abstract Algebra, Stochastic Analysis, Non-linear Dynamic, Mathematical Physics.
  • Statistics & Machine Learning: Prabability and Statistics, Multivariable Statistics, Machine Learning, Deep Learning.

Publications

• [1] Zhao, Yiming, Xiao Zhang, and Xinming Huang. “A Divide-and-Merge Point Cloud Clustering Algorithm for LiDAR Panoptic Segmentation.” In 2022 IEEE International Conference on Robotics and Automation (ICRA) 2022.
• [2] Zhao, Yiming, Xiao Zhang, and Xinming Huang. “A Technical Survey and Evaluation of Traditional Point Cloud Clustering Methods for LiDAR Panoptic Segmentation.” In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 1st Workshop on Traditional Computer Vision in the Age of Deep Learning (TradiCV) (2021).
• [3] Zhao, Yiming, Mahdi Elhousini, Ziming Zhang, and Xinming Huang. “Distance Transform Neural Network for LiDAR Depth Completion.” IEEE Transactions on Neural Networks and Learning Systems (TNNLS).
• [4] Zhao, Yiming, Lin Bai, and Xinming Huang. “FIDNet: LiDAR Point Cloud Semantic Segmentation with Fully Interpolation Decoding.” In 2021 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)
• [5] Zhao, Yiming, Xinming Huang, and Ziming Zhang. “Deep Lucas-Kanade Homography for Multimodal Image Alignment.” In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 15950-15959. 2021.
• [6] Zhao, Yiming, Lin Bai, Ziming Zhang, and Xinming Huang. “A Surface Geometry Model for LiDAR Depth Completion.” IEEE Robotics and Automation Letters 6 (RAL), no. 3 (2021): 4457-4464.
• [7] Bai, Lin, Yiming Zhao, and Xinming Huang. “3D Vehicle Detection Using Camera and Low-Resolution LiDAR.” arXiv preprint arXiv:2105.01765 (2021).
• [8] Bai, Lin, Yiming Zhao, and Xinming Huang. “A CNN accelerator on FPGA using depthwise separable convolution.” IEEE Transactions on Circuits and Systems II: Express Briefs (TCAS II)65, no. 10 (2018): 1415-1419.
• [9] Zhao, Yiming, An Zeng, Xiao-Yong Yan, Wen-Xu Wang, and Ying-Cheng Lai. “Unified underpinning of human mobility in the real world and cyberspace.” New Journal of Physics (NJP)18, no. 5 (2016): 053025.