Hongjun Jia's Email Addresses & Phone Numbers

📖 Summary

Instructor @ Switch Circuit Laboratory Lead and organize the class, demonstrate labs and grade on lab results and homework From January 2006 to June 2007 (1 year 6 months) Research associate @ The Ohio State University - Proposed a 3D object reconstruction algorithm with a novel Deviation Parameter criterion which measures the sub-dataset based on matrix perturbation theory and random matrix theory. Improved the performance of structure-from-motion algorithm significantly and handled up to 50% missing data successfully. Published several papers in the top journals and conferences in this field. Built a 3D mesh model to facilitate the verification.- Developed two new algorithms for Face Recognition and Pattern Classification: Support Vector Machines based algorithm and Partial Within-Class Match algorithm. Implemented them in Matlab and C++. Solved the problem of recognizing faces from a dataset with partial faces, which could be hardly resolved before. Extended the application scope of face recognition tremendously and achieved much higher recognition rate.- Participated in the team work of hand gesture / facial components detection and hand shape / facial expression analysis in the projects funded from NIH and NSF. - Constructed a video-based hand shape recognition system and applied it in a labyrinth game in Matlab and C. From September 2004 to March 2009 (4 years 7 months) Grader for Systems and Signals I @ The Ohio State University Grade homework From September 2005 to December 2005 (4 months) Research assistant @ Tsinghua University - Developed a series of new algorithms for motion estimation in H.264 video coding. These methods speeded up the inter-frame motion estimation process up to 55%. Published several papers in top journal and conferences which have been cited more than 30 times. Invited report on related topics was given in other universities in China.- Proposed a new one-pixel-wide edge detection method based on local accumulative histogram.- Worked on robust edge detection, image segmentation/classification, image enhancement and other image processing techniques on medical images, e.g. MRI and CT. From August 2001 to August 2004 (3 years 1 month) Embedded System Engineer @ Tsinghua Tongfang Co. Ltd Designed a multi-sources content indexing / searching system, an on-screen display system and the file download system on STi5512-based DVB-C Set Top Box.Tested and debugged codes from other groups. From May 2000 to August 2001 (1 year 4 months) Intern of Software Engineer @ Beijing Institute of Control Engineering Led a four person team on the prototype software for Electronic Components Simulation System with C / C++.Prepared part of the software-related documents as a project of the course of software engineering. From July 2000 to September 2000 (3 months) Software Engineer @ Amazon Prime Video Boston, Massachusetts, United StatesSenior Software Engineer @ Cognex Corporation – Working on software development on industry leading machine vision software, VisionPro, CVL, and the next generation of industrial 3D Vision Tools.– Developing fast and robust vision software for industries and researches for leading innovations for vision– Responsible for designing and developing software that solves the core machine vision problems for applications in logistics, robotic assembly, such as model extraction, defect detection, verification and measurement, pattern recognition and matching, spike removal, and calibration.– Developing machine vision guided robotics applications on Universal Robot UR10.– Two patents have been granted. From 2011 to 2020 (9 years) Greater Boston AreaPost-doc research fellow @ University of North Carolina at Chapel Hill - Working on several novel 3D/4D medical image registration, segmentation and classification algorithms, including groupwise registration, multi-atlas based segmentation, and disease related data mining and classification algorithms.- Designed and released software of ABSORB - an effective groupwise 3D image registration algorithm ((GUI and command line version)) to the public through NITRC, which has been extensively downloaded. - Published more than ten journal and international conference papers From 2009 to 2011 (2 years)