S. Alireza Golestaneh
About me
I am a Project Scientist at The Robotics Institute at Carnegie Mellon University.
Formerly I worked as a Postdoctoral Fellow at CMU with Prof. Kris Kitani since May 2019. Before that I spent one year at Human Sensing Lab working on video object detection while I was working with Dr. Dong Huang and Prof. Fernando de la Torre.
I received my Ph.D. in Electrical Engineering in 2018 from Arizona State University under the supervision of Prof. Lina Karam.
Before starting my Ph.D., I received my M.Sc. in Electrical Engineering in 2013 from Oklahoma State University under the supervision of Prof. Damon Chandler. I received my B.Sc. in Electrical Engineering under the supervision of Prof. Haidar Samet in 2010 from Shiraz University.
I also worked as an intern in Dolby Vision in Spring and Summer 2014 working under the supervision of Dr. Guan-Ming Su, at Dolby I also collaborated closely with Dr. Scott Daly.
My research interests are in computer vision, image processing, human vision understanding, visual perception, machine learning, and deep learning.
Interests
Computer Vision, Machine Learning, Perception
Selected Publications and Patents:
Importance of Self-Consistency in Active Learning for Semantic Segmentation
S. A Golestaneh, K Kitani | The British Machine Vision Conference (BMVC) | 2020
3D Human Motion Estimation via Motion Compression and Refinement
Z. Luo, S. A. Golestaneh, K. M. Kitani | 2020 (ACCV) | 2020
No-Reference Image Quality Assessment via Feature Fusion and Multi-Task Learning
S. A Golestaneh, K Kitani | 2020
[GitHub][pdf]
Synthesized Texture Quality Assessment via Multi-scale Spatial and Statistical Texture Attributes of Image and Gradient Magnitude Coefficients
S. A Golestaneh, LJ Karam | CVPRW | 2018
[GitHub][pdf] [project page]
Spatially-Varying Blur Detection Based on Multiscale Fused and Sorted Transform Coefficients of Gradient Magnitudes
S. A Golestaneh, LJ Karam | CVPR | 2017
[GitHub][pdf] [project page]
Reduced-Reference Quality Assessment Based on the Entropy of DWT Coefficients of Locally Weighted Gradient Magnitudes
SA Golestaneh, LJ Karam | IEEE Transactions on Image Processing | 2016
[GitHub][pdf] [project page]
Reduced-reference synthesized-texture quality assessment based on multi-scale spatial and statistical texture attributes
SA Golestaneh, LJ Karam | Image Processing (ICIP), 2016 IEEE International Conference on | 2016
Reduced-reference quality assessment based on the entropy of DNT coefficients of locally weighted gradients
SA Golestaneh, LJ Karam | Image Processing (ICIP), 2015 IEEE International Conference on | 2015
The effect of texture granularity on texture synthesis quality
SA Golestaneh, MM Subedar, LJ Karam | SPIE Optical Engineering+ Applications | 2015
[pdf] project page]
Scene-Change Detection Using Video Stream Pairs
S Golestaneh, GM Su, , GM Su | US Patent App. 14/801,633
Use of a local cone model to predict essential CSF light adaptation behavior used in the design of luminance quantization nonlinearities
S Daly, SA Golestaneh | Human Vision and Electronic Imaging | 2014
Algorithm for JPEG artifact reduction via local edge regeneration
SA Golestaneh, DM Chandler | Journal of Electronic Imaging | 2014
No-reference quality assessment of JPEG images via a quality relevance map
SA Golestaneh, DM Chandler | IEEE signal processing letters | 2013
Honors and Awards:
Nominated and Finalist for two (Sun Devil Spirit Award and Outstanding Graduate Student Leader) Pitchfork Awards 2017
Director of Wellness, Graduate and professional student association 2016-2017
Graduate Excellent Teaching Award, Arizona State University 2015-2016
University Graduate Fellowships, Arizona State University 2014-2015
Graduate Research Excellence Award, Oklahoma State University, 2014
Graduated in top 1% of my class, Oklahoma State University
Professional Activities:
Carnegie Mellon University: MSCV Admission Committee (2020)
Reviewer: TPAMI, TIP, TMM, JEI, CVPR, ICCV, BMVC
Assembly member of Graduate and professional student association (GPSA) 2015-2016
IEEE member 2010-Present