"Double Husky" aka Alumni of Northeastern University, Boston, MA
In a nutshell
Recently obtained PhD from SMILE Lab, NEU, and then joined the AI team as an engineer and researcher at Vicarious Surgical. Outside of work and academia, I enjoy catching up with family and friends, playing hockey or rollerblading with Jax (my dog), doing the occasional cook-off, traveling, skiing, and, more often than not, some light reading.
About Me [Google Scholar] [CURRICULUM VITAE (CV), below] Joseph P. Robinson received a BS in Electrical & Computer Engineering (2014) and a Ph.D. in Computer Engineering (2020) at Northeastern University (NEU), while also working as part-time faculty: designed & taught an undergrad course in Data Analytics. Research is in applied machine vision, emphasizing faces, deep learning, multimedia, and large databases. Previously, I led the team to TRECVid debut (MED, obtained 3rd best accuracy). I also built many image-based & video datasets, most notably Families In the Wild (FIW), which has recently extended with multimedia (i.e., FIW-Multimedia, aka FIW-MM, audio, video, and audio-visual). Balanced Faces in the Wild (i.e., BFW) is another face-based labeled dataset to support bias research in facial recognition [learn more here]. Has served as organizing chair and host of various workshops & challenges (e.g., NECV 2017, RFIW17@ACM-MM, RFIW18-RFIW19-RFIW20@FG, AMFG@CVPR18-19, FacesMM@ICME18-19), tutorials (ACM-MM18, FG19, CVPR19), PC member (e.g., CVPR, FG, MIRP, MMEDIA, AAAI, ICCV, ECCV, IJCAI), reviewer (e.g., IEEE Trans. on Biomedical Circuits and Systems, Image Processing, TPAMI, TIP), and leadership positions like the president of IEEE@NEU & Relations Officer of IEEE SAC R1 Region. Completed two NSF REUs (2010 & 2011); co-ops at Analogic Corporation & Raytheon BBN Technology; interned at MIT Lincoln Labs (2014), System & Technology Research (2016 & 2017), Snap Inc. (i.e., Snapchat) (2018), and ISMConnect (2019). Upon completing his Ph.D., Joseph joined Vicarious Surgical ASDAI group (Team Perception) as a full-time employee (2021).
Doctor Robinson's research interests include robust real-time facial recognition, deep learning, and human-computer interaction with a focus on the social implications of bringing computers into the medical room. He is particularly interested in exploring the social challenges of integrating computing into the day-to-day lives of surgeons – and how this will change the medical field, procedures, and practices.
Articles, a selection:
J. P. Robinson, M. Shao, and Y. Fu. "Survey on the Analysis and Modeling of Visual Kinship: A Decade in the Making." in IEEE Transactions on Pattern Analysis & Machine Intelligence (PAMI) 01 (2021).
J. P. Robinson, Y. Li, N. Zhang Y. Fu, and S Tulyakov. "Laplace Landmark Localization."Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 10103-10112
J. P. Robinson, M. Shao, Y. Wu, H. Liu, T. Gillis and Y. Fu, "Visual Kinship Recognition of Families in the Wild," in IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), vol. 40, no. 11, pp. 2624-2637 (2018).
[ 07/06/2021 ] Our SuperFront: From Low-resolution to High-resolution Frontal Face Synthesis in 2021 ACM MM!
[ 05/09/2021 ] Commencement! Officially completed PhD in Computer Engineering!
[ 04/15/2021 ] Big and exciting news for us here at Vicarious Surgical [ LinkedIn Post ]!
[ 03/21/2021 ] Our work in generative modeling using family photos to be published in 2021 ICME [ paper ].
[ 03/18/2021 ] IEEE PAMI accepted: check out our survey on kinship recognition technology [ paper ].
[ 03/15/2021 ] Joined AI team of Vicarious Surgical full-time [ link ].
To Recognizing Families In the Wild: A Machine Vision Tutorial presented at ACM-MM 2018 [ slides ]
Visual kinship recognition challenges remain open-- we try to provide platforms to compete and publish consistently (i.e., approximately CFP annually). However, competition portals are still active for registration and submissions for scoring. For challenges on Codalab (Task 1, Task 2, Task 3). Information on previous evaluations: RFIW 2017 I, RFIW 2018, RFIW 2019, andRFIW 2020; also, available on Kaggle! To learn more about visual kinship recognition, see a previous tutorial (e.g., here at ACM MM 2018). Finally, see project page for data downloads, paper pile, links to code, and more. FIW Database for (5) Kaggle! (1) publications;