Jan 2023 - Present
MACHINE LEARNING SW ENGINEER
Joined the REURC. More coming Soon.
Jan 2023 - Present
Lecturer for Machine Learning. More coming Soon.
03/2021 - 07/2022
Built depth perception for 3D reconstruction from images and videos. Used SOTA deep models and post-processing to estimate the depth and its confidence maps at the target fps and accuracy. Acquired labeled data to report ratings for our robot–data capture required learning ROS and Gazebo. Designed and collected both simulated and real datasets to train and fine-tune deep models via multi-task and transfer learning; then precisely measured system performance (i.e., characterized and find shortcomings).
2015 - 2020
The dissertation includes 35+ papers with 500+ citations and my involvement in the global community. As an ALERT DHS STEM Career Development & Research Fellow, I researched pattern recognition and data synthesis from various signal types. Cutting-edge work in big data for automatic face understanding by introducing image and video databases with novel deep learning techniques as benchmarks; hosted several data challenge workshops to promote and attract researchers to the problem. Topics span big data, landmark detection, generative modeling, learning with minimal or imbalanced data, bias in ML regarding its impact on society, and the construction of multi-modal databases.
2011 - 2014
Bachelor of Science
Served as an RA in the Optical Science Lab and NUCAR, including two summers at NHS REUs. Also, won department-wide Senior Capstone (ranked 1st) and the Freshman Remote Control Design Contest (1st place).
Joseph P Robinson, et al., "Balancing Biases and Preserving Privacy on Balanced Faces in the Wild." arXiv preprint arXiv:2103.09118 (2021). [paper]
Joseph P Robinson, et. al., "Families In Wild Multimedia (FIW-MM): A Multi-Modal Database for Recognizing Kinship." in IEEE Transaction on Multimedia (2021). [paper]
Joseph P Robinson, et al. "Survey on the Analysis and Modeling of Visual Kinship: A Decade in the Making." in IEEE Trans. on Pattern Analysis & Machine Intelligence (2021). [paper]
Joseph P Robinson et. al. "Face Recognition: Too Bias, or Not Too Bias?." in CVPR Workshop (2020). [paper]
Joseph P Robinson et al. "Recognizing Families In the Wild (RFIW): The 4th Edition." in IEEE Automatic Face and Gesture Recognition (2020). [paper]
Yu Yin, Joseph P Robinson, Yulun Zhang, Yun Fu. "Joint Super-Resolution and Alignment of Tiny Faces." in AAAI (2020). [paper]
SW Tools & Interfaces
MATLAB®, Python, GPGPU, Markdown, Pytorch, Tensorflow, Keras, JAVA, CUDA, Spark, Bash, C++, ROS, Gazebo, C, LATEX, SolidWorks®, AutoCAD®, Confluence, Notion
Understand, applied, implemented, and analyzed textbook algorithms; invented and published novel methods; comprehensive understanding of deep learning, classic approaches, and statistical modeling; data capture, acquisition, labeling, and evaluation; problems of bias, along with minimal data, and the means to mitigate (e.g., semi-supervision, self-supervision, adversarial learning)
HW Cameras/Video feeds, Depth sensors, Spectrum analyzer, cluster computing, Function generator, Servos, Photo-detector, Oscilloscope, Multi-meter, Pi, Board design, Solder, Optical alignment & specs.
Teaching, Mentoring, and Advising
Ability to learn complex topics