ResearchBlog

News from participating projects

Here lists the interesting papers/tools/research ideas I found recently. As I also post these references to my participating projects, the right sidebar show RSSs to these projects' site for these posts.


 VisGraphics
Computer Vision / Image processing
 HPCSubmission deadline

SIGGRAPH '15

 Jan. 19, 2015
IEEE Vis '15


 Mar. 21/Mar. 31, 2015
  ACM MM '15
 Mar. 31, 2015
   SC '15
Apr. 3/Apr. 10 --> Apr. 17, 2015
  ACM MM '15 short paper Apr. 30, 2015

PG '15
 May 8/May 15, 2015

SIGGRAPH Asia '15
 Jun. 2, 2015
   LDAV '15 Jun. 24, 2015
PacificVis '15

 Sep. 26, 2014
   IPDPS '15
Oct. 10/OCt. 17, 2014

I3D '15

 Oct. 21, 2014
  CVPR '15
 Nov. 14, 2014
EuroVis '15

 Nov. 30/Dec. 05, 2014

Related tutorials/papers about convergence issue of spectral clustering

posted May 6, 2015, 10:05 AM by Teng-Yok Lee   [ updated May 6, 2015, 10:08 AM ]

I am applying spectral clustering for graph partitioning, but ARPACK can fail to converge to compute the eigenvectors of the smallest 20 eigenvalues. Then I found the following tutorials/papers about the convergence issue:

On the convergence of spectral clustering on random samples: the normalized case.
von Luxburg, U., Bousquet, O., and Belkin, M.
In Proceedings of the 17th Annual Conference on Learning Theory (COLT) (pp. 457 – 471). 2004.
pdf
 
Limits of spectral clustering.
von Luxburg, U., Bousquet, O., and Belkin, M.
Advances in Neural Information Processing Systems (NIPS) 17 (pp. 857 – 864), 2005.
pdf

A Tutorial on Spectral Clustering
Ulrike von Luxburg
2007
arXiv:0711.0189

One important thing I learn is that from the tutorial on spectral cluster:

We have to make sure that the eigenvalues of L corresponding to the eigenvectors used in unnormalized spectral clustering are significantly smaller than the minimal degree in the graph.

More detail are in the tutorial. It also introduces 2 ways to normalize.


Textbooks for scientific visualization

posted May 2, 2015, 4:10 PM by Teng-Yok Lee   [ updated May 13, 2015, 8:41 PM ]



Authors/Book Note
Visualization in Medicine: Theory, Algorithms, and Applications
Bernhard Preim, Dirk Bartz
(Google)
 
The Visualization Handbook
Charles D. Hansen, Chris R. Johnson
Academic Press, 2005
(Google)
 
Introduction to Scientific Visualization
Helen Wright
Springer Science & Business Media, 2007
(Google)
 
Real Time Volume Graphics
K. Engel, M. Hadwiger, J. Kniss, C. Rezk, Salama, and D. Weiskopf
A.K. Peters, 2006
(Google)
This book is based on the author's tutorials for SIGGRAPH.
Isosurfaces: Geometry, Topology, and Algorithms
Rephael Wenger
CRC Press, Jun 24, 2013
(Google)
The author's publication page provides a sample chapter.
Flow visualization, 2nd Edition.
Wolgang Merzkirch
Academic Press, 1987
(Google, Elseiver)
 
Scientific Visualization: Techniques and Applications
K.W. Brodlie, L.A. Carpenter, R.A. Earnshaw, J.R. Gallop, R.J. Hubbold, A.M. Mumford, C.D. Osland, P. Quarendon
Springer Science & Business Media, 1992
(Google, Springer)
This book is based on a set of documents written for the AGOCG Workshop in UK, 1991.
An Introductory Guide to Scientific Visualization
Rae Earnshaw, Norman Wiseman.
Springer Science & Business Media, 1992
(Google, Springer)
 

Interesting visualization projects

posted Sep 30, 2014, 5:05 PM by Teng-Yok Lee   [ updated Apr 13, 2015, 6:41 PM ]

Real neural network


http://nxxcxx.github.io/Neural-Network/

VisNEST - Visualization of Simulated Neural Brain Activity
http://www.jara.org/en/research/jara-hpc/research/details/csg-immersive-vis/visnest-visualization-of-simulated-neural-brain-activity/

Artificial neural network

Opening the black box - data driven visualization of neural networks
F.-Y. Tzeng and Kwan-Liu Ma
Proceedings of IEEE Visualization 2005, pp.383,390, 23-28 Oct. 2005
pdf

Visualization of Artificial Neural Network with WebGL
MARKUS SPRUNCK
http://www.sw-engineering-candies.com/blog-1/experimental-visualization-of-artificial-neural-network-with-webgl

Visualizing and Understanding Convolutional Networks
Matthew D. Zeiler and Rob Fergus
In ECCV 2014
pdf

Misc

Inviso: Visualization Hadoop performance by Netflix.
https://github.com/Netflix/inviso

Visualizing MNIST: An Exploration of Dimensionality Reduction
https://colah.github.io/posts/2014-10-Visualizing-MNIST/

Several GraphCut papers to read

posted Sep 30, 2014, 8:29 AM by Teng-Yok Lee   [ updated Sep 30, 2014, 8:29 AM ]

These papers are the suggestion to-read list of the Coursera course Discrete Inference and Learning in Artificial Vision (instructors: Nikos Paragios and Pawan Kumar)

https://www.coursera.org/course/artificialvision

AuthorsTitleBook
Chaohui Wang, Nikos Komodakis, Nikos ParagiosMarkov Random Field modeling, inference & learning in computer vision & image understanding: A surveyComputer Vision and Image Understanding 117(11): 1610-1627 (2013)
Yuri Boykov and Vladimir KolmogorovAn Experimental Comparison of Min-Cut/Max-Flow Algorithms for Energy Minimization in VisionIEEE Transactions on Pattern Analysis and Machine Intelligence, 26(9):1124-1137 (2004)
Yuri Boykov and Olga VekslerGraph Cuts in Vision and Graphics: Theories and ApplicationsHandbook of Mathematical Models in Computer Vision, edited by Nikos Paragios, Yunmei Chen and Olivier Faugeras. Springer, 2006
Yuri Boykov, Olga Veksler and Ramin ZabihFast Approximate Energy Minimization via Graph CutsIEEE Transactions on Pattern Analysis and Machine Inteligence, 23(11): 1222-1239 (2001)
Vladimir KolmogorovConvergent Tree-reweighted Message Passing for Energy MinimizationIEEE Transactions on Pattern Analysis and Machine Intelligence 28(10): 1568-1583 (2006)
Vladimir Kolmogorov and Ramin ZabihWhat Energy Functions can be Minimized via Graph Cuts?IEEE Transactions on Pattern Analysis and Machine Inteligence, 26(2): 147-159 (2004)
Nikos Komodakis, Georgios Tziritas, Nikos ParagiosPerformance vs computational efficiency for optimizing single and dynamic MRFs: Setting the state of the art with primal-dual strategiesComputer Vision and Image Understanding 112(1), 14-29 (2008)
Nikos Komodakis, Nikos Paragios, Georgios TziritasMRF Energy Minimization and Beyond via Dual DecompositionIEEE Transactions on Pattern Analysis Machine Intelligence 33(3): 531-552 (2011)
Ben Taskar, Carlos Guestrin and Daphne KollerMax-Margin Markov NetworksProceedings of Advances in Neural Information Processing Systems: (2003)
Ioannis Tsochantaridis, Thorsten Joachims, Thomas Hofmann and Yasmine AltunLarge Margin Methods for Structured and Interdependent Output VariablesJournal of Machine Learning Research, 6:1453-1484 (2005)
Tomas WernerRevisiting the Linear Programming Relaxation Approach to Gibbs Energy Minimization and Weighted Constraint SatisfactionIEEE Transactions on Pattern Analysis and Machine Intelligence 32(8): 1474-1488 (2010)


Recent work on image-based approaches for scientific visualization

posted Sep 29, 2014, 9:42 PM by Teng-Yok Lee   [ updated Oct 3, 2014, 7:05 PM ]

Explorable images for visualizing volume data.
Anna Tikhonova, Carlos D. Correa, Kwan-Liu Ma.
In Proceedings of IEEE Pacific Visualization Symposium 2010, pp. 177-184, 2010.

An Exploratory Technique for Coherent Visualization of Time-varying Volume Data.
Anna Tikhonova, Carlos D. Correa, Kwan-Liu Ma.
Computer Graphics Forum, 29(3):783-792, 2010.
pdf

Visualization by Proxy: A Novel Framework for Deferred Interaction with Volume Data.

Anna Tikhonova, Carlos D. Correa, Kwan-Liu Ma.
IEEE Transactions on Visualization and Computer Graphics, 16(6):1551-1559, 2010.

An Image-based Approach to Extreme Scale In Situ Visualization and Analysis.
James Ahrens, John Patchett, Sebastien Jourdain, David H. Rogers, Patrick O’Leary, and Mark Petersen.
SuperComputing 2014, to appear.

PS. Another set of relevant paper is to evaluate the visibility of a single view point:

Visibility-Driven Transfer Functions.
Carlos D. Correa, Kwan-Liu Ma.
In Proceedings of IEEE Pacific Visualization Symposium 2009, 2009.

Visibility Histograms and Visibility-Driven Transfer Functions.
Carlos D. Correa, Kwan-Liu Ma.
IEEE Transactions on Visualization and Computer Graphics, 17(2):192-204, 2011.

[FW] Common Misconceptions about Data Analysis and Statistics

posted Sep 27, 2014, 10:12 AM by Teng-Yok Lee

An article to refresh my memory on statistics?

Motulsky HJ, Common Misconceptions about Data Analysis and Statistics. J Pharmacol Exp Ther. 351(1):200-5, 2014 Oct.
doi: 10.1124/jpet.114.219170.
http://jpet.aspetjournals.org/content/351/1/200.full.pdf

IEEE Vis '14 papers to read

posted Sep 7, 2014, 11:07 PM by Teng-Yok Lee   [ updated Sep 24, 2014, 5:25 PM ]

REF: http://ieeevis.org/year/2014/info/overview-amp-topics/accepted-papers

SciVis

Characterizing Molecular Interactions in Chemical Systems
David Guenther, Roberto Alvarez Boto, Julia Contreras Garcia, Jean-Philip Piquemal, Julien Tierny

Vortex Cores of Inertial Particles
Tobias Günther, Holger Theisel

Advection-Based Sparse Data Management for Visualizing Unsteady Flow
Hanqi Guo, Jiang Zhang, Richen Liu, Lu Liu, Xiaoru Yuan, Jian Huang, Xiangfei Meng, Jingshan Pan

FLDA: Latent Dirichlet Allocation Based Unsteady Flow Analysis
Fan Hong, Chufan Lai, Hanqi Guo, Xiaoru Yuan, Enya Shen, Sikun Li

Fixed-Rate Compressed Floating-Point Arrays
Peter Lindstrom

Sparse PDF Volumes for Consistent Multi-Resolution Volume Rendering
Ronell Sicat, Jens Krueger, Torsten Möller, Markus Hadwiger
preprint

TVCG
Interpolation-Based Pathline Tracing in Particle-Based Flow Visualization
Jennifer Chandler, Harald Obermaier, Ken Joy

My notes on Logistic Regression

posted Jun 9, 2014, 10:32 PM by Teng-Yok Lee   [ updated Jun 10, 2014, 8:59 AM ]

REF: http://www.holehouse.org/mlclass/06_Logistic_Regression.html

Because it took me a while to finally derive it, I decide to put the detail here. Since I cannot type the equation nice, I simplify the notations.

Note 1: Gradients of logistic cost functions

Here the cost function is denoted as F(t) where t terms for theta. m is the number of samples. (x(i), y(i)), i = 1 ... m, is the training set. ht(x) =1/(1+exp(-tTx)) is the logistic function with parameter theta (t).

F(t) = 1/m sumi = 1 ... m y(i) log ht(x(i)) + (1 - y(i))log (1 - ht(x(i)))


The partial gradient w.tr.t tj is denoted as d/dtj = dj. Then the partial gradient at tj, aka, dj F(t) is derived as follows:

dj F(t
= 1/m sumi = 1 ... m y(i) / ht(x(i)) dj ht(x(i)) + (1 - y(i))/(1 - ht(x(i))) dj (1 - ht(x(i)))
= 1/m sumi = 1 ... m y(i) / ht(x(i)) dj ht(x(i)) -  (1 - y(i))/(1 - ht(x(i))) dj ht(x(i)) <-- Remove the constant 1, which is in boldface above.
= 1/m sumi = 1 ... m dj ht(x(i)) {y(i) / ht(x(i))  -  (1 - y(i))/(1 - ht(x(i)))}  <-- Separate dj ht(x(i)) from both terms.
= 1/m sumi = 1 ... m dj ht(x(i)) {y(i) (1 - ht(x(i))) - ht(x(i))(1 - y(i))}/{ht(x(i))(1 - ht(x(i))} 
= 1/m sumi = 1 ... m dj ht(x(i)) (y(i) - ht(x(i)))/{ht(x(i))(1 - ht(x(i))} 


where
dj ht(x(i))
= dj (1 + exp(-tTx(i)))-1
= -(1 + exp(-tTx(i)))-2 exp(-tTx(i)) x(i)j
= -{1/(1 + exp(-tTx(i))} {exp(-tTx(i)) / (1 + exp(-tTx(i))} x(i)j
= -{ht(x(i))(1 - ht(x(i))} x(i)j


Thus

dj F(t)
= 1/m sumi = 1 ... m dj ht(x(i)) (y(i) - ht(x(i)))/{ht(x(i))(1 - ht(x(i))}
= 1/m sumi = 1 ... m (y(i) - ht(x(i))) {-{ht(x(i))(1 - ht(x(i))} x(i)j}/{ht(x(i))(1 - ht(x(i))}
= 1/m sumi = 1 ... m (ht(x(i)) - y(i)) x(i)j

Note 2: How is logistic regression related to MLE?

Actually the logistic regression cost can be treated as the likelihood of a Bernoulli random variable. Here P[x; t] = ht(x) is the probability that x belongs to class 0. Then the pdf of x is
f(x) = P[x; t]y (1 - P[x; t])1 - y

and its likelihood of theta t is:
L(t) = log f(x) = y log P[x; t]+ (1- y) log 1 - P[x; t] = y log ht(x) + (1- y) log 1 - ht(x)
That's why optimizing the cost function F is equivalent to find the Maximum Likelihood Estimator.

Visualization of MPI/Parallel Programs

posted Jun 4, 2014, 10:50 PM by Teng-Yok Lee

Research articles


Visualizing Large-scale Parallel Communication Traces Using a Particle Animation Technique
Carmen Sigovan, Chris Muelder, and Kwan-Liu Ma
In Proceedings of EuroVis 2013.
pdf

Gantt Chart Visualization for MPI and Apache Multi-Dimensional Trace Files
Wu, C.E. and Bolmarcich, A.
In IPDPS 2002.
pdf

Software packages




Video Visualization

posted Dec 20, 2013, 10:03 PM by Teng-Yok Lee   [ updated Feb 27, 2014, 6:41 PM ]

A Survey on Video-based Graphics and Video Visualization
Rita Borgo, Min Chen, Ben Daubney, Edward Grundy, Gunther Heidemann, Benjamin Höferlin, Markus Höferlin, Heike Jänicke, Daniel Weiskopf and Xianghua Xie.
http://www.vis.uni-stuttgart.de/~weiskopf/publications/eg11_star.pdf

Video Visualization
http://www.oerc.ox.ac.uk/projects/video-visualization

I am especially interested techniques that create effective overviews for videos:

Techniques for interactive video cubism.
S. S. Fels, E. Lee, and K. Mase.
In MM ’00: Proceedings of ACM Multimedia 2000, pages 368–370, 2000.

Video visualization.
G. Daniel and M. Chen.
In VIS ’03: Proceedings of the IEEE Visualization 2003, pages 409–416, 2003.

Computational time-lapse video.
Eric P. Bennett and Leonard McMillan.
In SIGGRAPH 2007, Article 102, 2007.

Factored time-lapse video.
Kalyan Sunkavalli, Wojciech Matusik, Hanspeter Pfister, and Szymon Rusinkiewicz.
In SIGGRAPH 2007, Article 101, 2007.

Exploring video streams using slit-tear visualizations.
A. Tang, S. Greenberg, and S. Fels.
In AVI ’08: Proceedings of the Working Conference on Advanced Visual Interfaces 2008, pages 191–198, 2008.

Action-based multifield video visualization.
R. P. Botchen, S. Bachthaler, F. Schick, M. Chen, G. Mori, D. Weiskopf, and T. Ertl.
IEEE Transactions on Visualization and Computer Graphics 14(4):885–899, 2008.

Dynamic video narratives.
Carlos D. Correa and Kwan-Liu Ma.
In SIGGRAPH '10, Article 88 , 9 pages, 2010.

1-10 of 74