myPhoto
Chetan Bhole's Website
meliora
URCS

 

Welcome to my website!

These are a few pages about me and my research interests and work.

I am from Thane, a city on the outskirts of Mumbai in India. I have done my Bachelors in Computer Engineering at Fr. CRCE, Mumbai University and my Masters in Computer Science and Engineering at the University at Buffalo. I graduated from the Department Of Computer Science in University of Rochester (UR).

 

        Office Room no.
        Computer Science Building
        
        Phone:
        Email:
        last name at cs.rochester.edu

      Research

 

I currently work at A9 (an Amazon company) on product search,ranking and relevance.

My primary research interests during graduate school were in Computer Vision and Machine Learning, working on graphical models (Conditional Random Fields in particular) for image segmentation. I have used CT image volumetric data for medical image segmentation and more recently have been looking into use of videos. My co-advisors are Dr. Chris Pal and Dr. Henry Kautz.

I interned at Google in the summer of 2012 working of matching images to YouTube videos with the Machine perception team in Google Research and also had worked the summer before in 2011 on full body human pose estimation with the Google Products team.

I spent the summer of 2008 as an intern at Carestream Health Inc. where I worked on medical document classification using different learning models.

I was previously involved in a Computer Aided Diagnosis project (Segmentation of vertebral columns from MR scans) at University at Buffalo (UB) under the guidance of Dr. Vipin Chaudhary. During my Masters at UB, I worked under Dr. Peter Scott on object recognition. I was also involved as a research assistant in projects related to document analysis in CEDAR and image reconstruction at the Department of Nuclear Medicine at UB.

View Chetan Bhole's profile on LinkedIn

 

      Code

 

  • Image and video segmentation using CRF (Conditional Random Fields) (C++ code)
    [Project Page]

  • Dual Decomposition for inference (C++)
    [Project Page]

      Publications, Reports and Posters

 

blank Approximate inference using unimodular graphs in dual decomposition
C. Bhole, J. Domke, D. Gildea
Optimization in Machine Learning, NIPS workshop, 2013
Paper [link] Supplement [pdf]

blank 3D segmentation of abdominal CT imagery with graphical models, conditional random fields and learning
C. Bhole, C. Pal, D. Rim, A. Weismuller
Machine Vision and Applications, pp. 1-25 (Springer Online First Article) 2013
Paper [link]

tennis Automated person segmentation in videos
C. Bhole, C. Pal
In proceedings of ICPR 2012
Paper [link]
[Project Page]

blank 3D Segmentation in CT Imagery with Conditional Random Fields and Histograms of Oriented Gradients
C. Bhole, N. Morsillo, C. Pal
Machine Learning in Medical Imaging, MICCAI 2011 Workshop
Paper [pdf]. Appendix [pdf]
[Data Details]

myPhoto Context Sensitive Labeling of Spinal Structure in MR Images
C. Bhole, S. Kompalli, V. Chaudhary
Poster presentation, Conference Proceeding at SPIE Medical Imaging, 2009.
Paper [pdf]. Poster [pdf]

myPhoto Object recognition using shape and behavioral features
C. Bhole, Advisor: Peter D. Scott
Masters Thesis at SUNY UB, 2006.
Report [pdf]

myPhoto Parallel OSEM reconstruction speed with MPI, OpenMP and Hybrid MPI-OpenMP programming models
M. Jones, R. Yao, C. Bhole
IEEE Transactions on Nuclear Science, vol. 53, pp. 2752-2758, 2006.
Paper [pdf]

myPhoto Spotting words in handwritten Arabic documents
S. Srihari, H. Srinivasan, P. Babu, C. Bhole
Conference Proceeding at IS&T/SPIE Electronic Imaging 2006 Symposium.
Paper [pdf]

myPhoto Handwritten Arabic Word Spotting using the CEDARABIC Document Analysis System
S. Srihari, H. Srinivasan, P. Babu, C. Bhole
Proc. Symposium on Document Image Understanding (SDIUT 05), College Park, MD, November 2005.
Paper [pdf]