Kun He | BUCS


PhD candidate, Computer Science, Boston University

Curriculum Vitae (updated Feb 2018)


PhD candidate in computer science, member of Image and Video Computing Group, advised by Prof. Stan Sclaroff. I obtained my M.Sc. degree in computer science from Boston University in 2013, and my B.Sc. degree (with honors) in computer science from Zhejiang University in 2010.

I'm interested in exploring solutions to computer vision problems using machine learning techniques, and I'm frequently attracted by the optimization aspect in such problems. I have worked on topics in object detection, pose estimation, and image segmentation. Currently my focus is on learning vector embeddings for high-dimensional images/videos, which find application in image retrieval and other related vision tasks. The formulations being developed cover both binary embeddings (aka. learning to hash) and continuous embeddings (aka. metric learning).

I did a few research internships: with Dr. Yan Lu at Honda Research Institute during Summer 2017, and with Dr. Leonid Sigal at Disney Research Pittsburgh during Summer 2015 and Fall 2013. In Spring 2014, I was a visiting student in Prof. Pedro Felzenszwalb's group at Brown University.



Google Scholar profile

Journal Publications


Predicting Foreground Object Ambiguity and Efficiently Crowdsourcing the Segmentation(s) [arXiv] [Project]
Danna Gurari, Kun He, Bo Xiong, Jianming Zhang, Mehrnoosh Sameki, Suyog Dutt Jain, Stan Sclaroff, Margrit Betke, Kristen Grauman
IJCV, 2018


Hashing with Mutual Information [arXiv]
Kun He*, Fatih Cakir*, Sarah Adel Bargal, Stan Sclaroff
(* Equal contribution)
Under review, IEEE TPAMI

Conference Publications


Local Descriptors Optimized for Average Precision
Kun He, Yan Lu, Stan Sclaroff
CVPR, 2018 (to appear)


Hashing as Tie-Aware Learning to Rank [arXiv]
Kun He, Fatih Cakir, Sarah Adel Bargal, Stan Sclaroff
CVPR, 2018 (to appear)


MIHash: Online Hashing with Mutual Information [PDF] [poster] [code]
Kun He*, Fatih Cakir*, Sarah Adel Bargal, Stan Sclaroff
(* Equal contribution)
ICCV, 2017


Parameterizing Object Detectors in the Continuous Pose Space [PDF] [BibTeX] [poster] [video]
Kun He, Leonid Sigal, Stan Sclaroff
ECCV, 2014
[Tech report on implementation]


Scale Resilient, Rotation Invariant Articulated Object Matching [PDF] [BibTeX]
Hao Jiang, Tai-Peng Tian, Kun He, Stan Sclaroff
CVPR, 2012



Generalized Majorization-Minimization [arXiv]
Sobhan Naderi Parizi, Kun He, Stan Sclaroff, Pedro Felzenszwalb
arXiv pre-print, 2015


Stochastic Functional Descent for Learning Support Vector Machines [PDF]
Kun He
M.Sc. thesis, 2013