I am a PhD student at UT Austin, where I have been working with my advisor Prof. Kristen Grauman since Spring '13.
I joined UT in Fall 2011 on an MCD Fellowship. Before this, I obtained my bachelor's degree at Indian Institute of Technology, Madras (IITM), where I majored in electrical engineering and minored in literature.
I spent the summer of 2014 at UC Berkeley, working with Prof. Alyosha Efros on cross-view geolocation. In 2012, I summer-interned at Intel Labs, Santa Clara, working on computational photography techniques with camera arrays. During my undergrad, I interned in 2010 at Marvell Semiconductors, Bangalore where I worked on video quality analysis. I was also briefly a graduate research assistant at LIVE lab in Fall 2012.
- Nov 2016: Our paper on automatic active cinematography received the Best Application Paper Award at ACCV 2016!
- Sep 2016: Videos from our ECCV workshop are posted here.
- Aug 2016: Models for 227x227 images based on our ICCV 2015 egomotion-equivariance paper released.
- Two new papers on active cinematography and unsupervised feature learning, coming up at ACCV 2016, Taipei.
- New paper on end-to-end active vision, coming up at ECCV 2016, Amsterdam.
- Together with my advisor Kristen Grauman, and Sergey Levine over at UW, I'm organizing the 1st Workshop on Action and Anticipation for Visual Learning at ECCV 2016, Amsterdam in October this year, with an exciting, interdisciplinary focus.
- Mar 2016: Thanks to Samsung for generously funding me through a Samsung USA PhD Fellowship for 2016-17!
- Computer vision, specifically visual recognition
- Machine learning
- Computational photography
- Image processing
Google Scholar profile] [Semantic Scholar profile] [Microsoft Academic profile] [DBLP profile]
- Dinesh Jayaraman and Kristen Grauman, Learning image representations tied to egomotion from unlabeled video, Invited Paper in IJCV 2017 Special Issue of Best Papers from ICCV 2015. [Published version (ReadCube)] [Downloadable preprint] [Project page with codes, models, additional results]
- Yu-Chuan Su, Dinesh Jayaraman and Kristen Grauman, Pano2Vid: Automatic cinematography for watching 360-degree videos, (oral presentation at) ACCV 2016, Taipei, November 2016 (Best Application Paper Award) [main pdf] [supp pdf]
- Ruohan Gao, Dinesh Jayaraman and Kristen Grauman, Object-Centric Representation Learning from Unlabeled Videos, ACCV 2016, Taipei, November 2016 [Project page] [main pdf]
- Dinesh Jayaraman and Kristen Grauman, Look-ahead before you leap: end-to-end active vision by forecasting the effect of motion, (oral presentation at) ECCV 2016, Amsterdam, Netherlands, October 2016 [Project Page] [main pdf] [supp pdf] [UTCS Tech Report (Dec'15)] [arXiv] [ECCV'16 oral slides] [ECCV'16 poster] [oral video]
- Dinesh Jayaraman and Kristen Grauman, Slow and steady feature analysis: higher order temporal coherence in video, (spotlight presentation at) CVPR 2016, Las Vegas, NV, June 2016 [main pdf] [supp pdf] [arXiv (June 2015 version)] [CVPR'16 spotlight slides] [CVPR'16 poster] [spotlight video]
- Dinesh Jayaraman*, Chao-Yeh Chen* and Kristen Grauman(* indicates equal contribution), Divide, Share, and Conquer: Multi-task Attribute Learning with Selective Sharing, Springer book on visual attributes (Editors: Rogerio Feris, Devi Parikh, Christoph H. Lampert), 2016 [main pdf]
- Dinesh Jayaraman and Kristen Grauman, Learning image representations tied to ego-motion, (oral presentation at) ICCV 2015, Santiago, Chile, December 2015 [Project Page] [main pdf] [supp pdf] [ICCV'15 oral slides] [ICCV'15 poster] [arXiv] [oral video] (also appeared at Object Understanding and Interaction Workshop, ICCV '15)
- Dinesh Jayaraman and Kristen Grauman, Zero-shot recognition with unreliable attributes, NIPS 2014, Montreal, Canada, December 2014 [Project Page] [main pdf] [supp pdf] [arXiv] (also appeared at Language and Vision Workshop, CVPR'15)
- Dinesh Jayaraman, Fei Sha and Kristen Grauman, Decorrelating semantic visual attributes by resisting the urge to share, (oral presentation at) CVPR 2014, Columbus OH, June 2014 [Project Page] [main pdf] [supp pdf] [CVPR'14 oral slides] [CVPR'14 poster] [oral video] (also appeared at Parts and Attributes Workshop, ECCV '14)
- Dinesh Jayaraman, Anish Mittal, Anush Moorthy and Alan Bovik, Objective quality assessment of multiply distorted images, Asilomar Conference on Signals and Systems, Pacific Grove CA, October 2012 [Project Page] [pdf]
- Dinesh Jayaraman, Tao Ma, Wei Sun, Oscar Nestares and Kalpana Seshadrinathan, Techniques for rectification of camera arrays (US patent application submitted November 2012)
- Dinesh Jayaraman, Oscar Nestares, Kalpana Seshadrinathan, Techniques for improved focusing of camera arrays (US patent application submitted March 2013)
- CS 395T Visual Recognition
- PSY 380E Vision Systems
- CS 395T Graphical Models
- EE 381V Large-Scale Learning
- EE 381K Information Theory
- EE 381J Probability and Stochastic Processes
- EE 381V Game Theory
- ORI 391Q Linear Programming
- EE 381V Multi-scale/rate signal processing
- EE 381K Digital Signal Processing
- CS 391L Machine Learning
- CS 669 Multi-View Geometry in Vision*
- CS 669 Pattern Recognition*
- CS 631 Artificial Neural Networks*
My desk is at GDC (Gates and Dell Complex) 4.728 E on the UT main campus. In most cases, I am best reached at: dineshj [at] cs [dot] utexas [dot] edu.
- Bo Xiong
- Yu-Chuan Su
- Yong Jae
- Sung Ju
- Lu Zheng
My academic genealogy at Neurotree.