Importance-Driven Time-Varying Data Visualization Chaoli Wang, Hongfeng Yu, Kwan-Liu Ma University of California, Davis.

Slides:



Advertisements
Similar presentations
Dynamic View Selection for Time-Varying Volumes Guangfeng Ji* and Han-Wei Shen The Ohio State University *Now at Vital Images.
Advertisements

LOD Map – A Visual Interface for Navigating Multiresolution Volume Visualization Chaoli Wang and Han-Wei Shen The Ohio State University Presented at IEEE.
Cheuk Yiu Ip Amitabh Varshney Joseph JaJa
Feature Selection as Relevant Information Encoding Naftali Tishby School of Computer Science and Engineering The Hebrew University, Jerusalem, Israel NIPS.
Image Repairing: Robust Image Synthesis by Adaptive ND Tensor Voting IEEE Computer Society Conference on Computer Vision and Pattern Recognition Jiaya.
Carolina Galleguillos, Brian McFee, Serge Belongie, Gert Lanckriet Computer Science and Engineering Department Electrical and Computer Engineering Department.
An Information-Theoretic Framework for Flow Visualization Lijie Xu, Teng-Yok Lee, & Han-Wei Shen The Ohio State University.
Segmentation via Maximum Entropy Model. Goals Is it possible to learn the segmentation problem automatically? Using a model which is frequently used in.
Lecture 3 Nonparametric density estimation and classification
Christine Smyth and Jim Mori Disaster Prevention Research Institute, Kyoto University.
Computer Vision Group, University of BonnVision Laboratory, Stanford University Abstract This paper empirically compares nine image dissimilarity measures.
1 Hierarchical Image-Motion Segmentation using Swendsen-Wang Cuts Adrian Barbu Siemens Corporate Research Princeton, NJ Acknowledgements: S.C. Zhu, Y.N.
EE 7730 Image Segmentation.
Prénom Nom Document Analysis: Data Analysis and Clustering Prof. Rolf Ingold, University of Fribourg Master course, spring semester 2008.
Dimension reduction : PCA and Clustering by Agnieszka S. Juncker
Multi-Scale Analysis for Network Traffic Prediction and Anomaly Detection Ling Huang Joint work with Anthony Joseph and Nina Taft January, 2005.
Project 4 out today –help session today –photo session today Project 2 winners Announcements.
Gaussian Information Bottleneck Gal Chechik Amir Globerson, Naftali Tishby, Yair Weiss.
Dimension reduction : PCA and Clustering Christopher Workman Center for Biological Sequence Analysis DTU.
Knowledge-Assisted Visualization of Turbulent Combustion Simulations Chaoli Wang, Hongfeng Yu, Kwan-Liu Ma.
Lecture #1COMP 527 Pattern Recognition1 Pattern Recognition Why? To provide machines with perception & cognition capabilities so that they could interact.
Social-Aware Collaborative Visualization for Large Scientific Projects Kwan-Liu Ma and Chaoli Wang CTS’085/21/2008.
Computer Vision I Instructor: Prof. Ko Nishino. Today How do we recognize objects in images?
Pattern Recognition. Introduction. Definitions.. Recognition process. Recognition process relates input signal to the stored concepts about the object.
Sufficient Dimensionality Reduction with Irrelevance Statistics Amir Globerson 1 Gal Chechik 2 Naftali Tishby 1 1 Center for Neural Computation and School.
Review Rong Jin. Comparison of Different Classification Models  The goal of all classifiers Predicating class label y for an input x Estimate p(y|x)
A Multiresolution Volume Rendering Framework for Large-Scale Time- Varying Data Visualization Chaoli Wang 1, Jinzhu Gao 2, Liya Li 1, Han-Wei Shen 1 1.
Image Pyramids and Blending
Methods in Medical Image Analysis Statistics of Pattern Recognition: Classification and Clustering Some content provided by Milos Hauskrecht, University.
In Situ Sampling of a Large-Scale Particle Simulation Jon Woodring Los Alamos National Laboratory DOE CGF
EuroVis 2013 The Eurographics Conference on Visualization Evaluating Isosurfaces with Level-set based Information Maps Tzu-Hsuan Wei, Teng-Yok Lee, and.
The MPEG-7 Color Descriptors
Venkatram Ramanathan 1. Motivation Evolution of Multi-Core Machines and the challenges Summary of Contributions Background: MapReduce and FREERIDE Wavelet.
1 Logistic Regression Adapted from: Tom Mitchell’s Machine Learning Book Evan Wei Xiang and Qiang Yang.
FlowString: Partial Streamline Matching using Shape Invariant Similarity Measure for Exploratory Flow Visualization Jun Tao, Chaoli Wang, Ching-Kuang Shene.
Texture. Texture is an innate property of all surfaces (clouds, trees, bricks, hair etc…). It refers to visual patterns of homogeneity and does not result.
Image Registration as an Optimization Problem. Overlaying two or more images of the same scene Image Registration.
Information Theory Basics What is information theory? A way to quantify information A lot of the theory comes from two worlds Channel.
A New Method of Probability Density Estimation for Mutual Information Based Image Registration Ajit Rajwade, Arunava Banerjee, Anand Rangarajan. Dept.
Lei Zhang and Guoning Chen, Department of Computer Science, University of Houston Robert S. Laramee, Swansea University David Thompson and Adrian Sescu,
Efficient Local Statistical Analysis via Integral Histograms with Discrete Wavelet Transform Teng-Yok Lee & Han-Wei Shen IEEE SciVis ’13Uncertainty & Multivariate.
Generalized Model Selection For Unsupervised Learning in High Dimension Vaithyanathan and Dom IBM Almaden Research Center NIPS ’ 99.
Mining Anomalies Using Traffic Feature Distributions Anukool Lakhina Mark Crovella Christophe Diot in ACM SIGCOMM 2005 Presented by: Sailesh Kumar.
Image Segmentation Shengnan Wang
Designing Factorial Experiments with Binary Response Tel-Aviv University Faculty of Exact Sciences Department of Statistics and Operations Research Hovav.
Xin Tong, Teng-Yok Lee, Han-Wei Shen The Ohio State University
CS654: Digital Image Analysis
Feature Selction for SVMs J. Weston et al., NIPS 2000 오장민 (2000/01/04) Second reference : Mark A. Holl, Correlation-based Feature Selection for Machine.
Scientific Data Analysis via Statistical Learning Raquel Romano romano at hpcrd dot lbl dot gov November 2006.
ITree: Exploring Time-Varying Data using Indexable Tree Yi Gu and Chaoli Wang Michigan Technological University Presented at IEEE Pacific Visualization.
1 C.A.L. Bailer-Jones. Machine Learning. Data exploration and dimensionality reduction Machine learning, pattern recognition and statistical data modelling.
Rendering Pipeline Fall, 2015.
Distributions cont.: Continuous and Multivariate
Dimension reduction : PCA and Clustering by Agnieszka S. Juncker
Project Implementation for ITCS4122
6. Introduction to nonparametric clustering
Image Segmentation Techniques
Lecture 26: Faces and probabilities
Homogeneity Guided Probabilistic Data Summaries for Analysis and Visualization of Large-Scale Data Sets Ohio State University (Shen) Problem: Existing.
Efficient Distribution-based Feature Search in Multi-field Datasets Ohio State University (Shen) Problem: How to efficiently search for distribution-based.
Data Transformations targeted at minimizing experimental variance
Announcements Project 2 artifacts Project 3 due Thursday night
Announcements Project 4 out today Project 2 winners help session today
Announcements Artifact due Thursday
J.M. Sotoca, F. Pla, A. C. Klaren
Announcements Artifact due Thursday
Spatial statistics of X-ray volumes reveal layering and spatially diverse distribution of cell bodies. Spatial statistics of X-ray volumes reveal layering.
Feature computation and classification of grating pitch.
The “Margaret Thatcher Illusion”, by Peter Thompson
Time-varying volume visualization and compression
Presentation transcript:

Importance-Driven Time-Varying Data Visualization Chaoli Wang, Hongfeng Yu, Kwan-Liu Ma University of California, Davis

Importance-Driven Volume Rendering [Viola et al. 04]

Differences Medical or anatomical data sets Pre-segmented objects Importance assignment Focus on rendering Time-varying scientific data sets No segmentation or objects are given Importance measurement Focus on data analysis

Questions How to capture the important aspect of data? Importance – amount of change, or “unusualness” How to utilize the importance measure? Data classification Abnormality detection Time budget allocation Time step selection

Related Work Time-varying data visualization Spatial and temporal coherence [Shen et al. 94, Westermann 95, Shen et al. 99] Compression, rendering, presentation [Guthe et al. 02, Lum et al. 02, Woodring et al. 03] Transfer function specification [Jankun-Kelly et al. 01, Akiba et al. 06] Time-activity curve (TAC) [Fang et al. 07] Local statistical complexity (LSC) [Jänicke et al. 07]

Importance Analysis Block-wise approach Importance evaluation Amount of information a block contains by itself New information w.r.t. other blocks in the time series Information theory Entropy Mutual information Conditional entropy

Information Theory Entropy Mutual information Conditional entropy p(x), p(y) : Marginal probability distribution function p(x,y) : Joint probability distribution function

Relations with Venn Diagram H(X)H(X)H(Y)H(Y) I(X;Y)I(X;Y)H(X|Y)H(X|Y)H(Y|X)H(Y|X) I(X;Y) = I(Y;X)H(X|Y) ≠ H(Y|X)

Entropy in Multidimensional Feature Space Feature vector Data value Gradient magnitude or other derivatives Domain-specific quantities Multidimensional histogram Use the normalized bin count as probability p(x) f1f1 f3f3 f2f2

Importance in Joint Feature-Temporal Space Consider two data blocks X and Y at the same spatial location neighboring time steps Use joint feature-temporal histogram Use the normalized bin count as probability p(x,y) Run-length encode the histogram F F F = (f 1, f 2, f 3, …)

Importance Value Calculation Consider a time window for neighboring blocks Importance of a data block X j at time step t : Importance of time step t :

Importance Curve – Earthquake Data Set T I regular

Importance Curve – Climate Data Set T I periodic

Importance Curve – Vortex Data Set T I turbulent

Clustering Importance Curves Hybrid k -means clustering [Kanungo et al. 02] Lloyd’s algorithm Local search by swapping centroids Avoid getting trapped in local minima

Clustering All Time Steps vs. Time Segments 599 time steps 50 segments 1200 time steps 120 segments 90 time steps 90 segments

Cluster Highlighting – Earthquake Data Set

Cluster Highlighting – Hurricane Data Set

Cluster Highlighting – Climate Data Set

Cluster Highlighting – Vortex Data Set

Cluster Highlighting – Combustion Data Set

Abnormality Detection A: El NiñoB: La Niña

Time Budget Allocation Allocate time budget based on importance value Animation time Non-even allocation Rendering time Assign to each time step (and each block in a time step) Adjust the sampling spacing accordingly

Time Step Selection Uniform selection Importance-driven selection Select the first time step Partition the rest of time steps into (K-1) segments In each time segment, select one time step: Maximize the joint entropy

Precomputation and Clustering Performance The test data sets with their parameter settings, sizes of joint feature-temporal histograms, and timings for histogram calculation. Timing for clustering all time steps of the five test data sets.

Choices of Window and Bin Sizes The importance curve of the vortex data set with different time window sizes ( W ) and numbers of bins for feature components F = ( f 1, f 2, f 3 ).

Choices of # of Clusters and Block Size The cluster of the highest importance values under different choices of number of clusters and block size. Top row: color adjustment only. Bottom row: color and opacity adjustment. 3 clusters4 clusters5 clusters 50×50×2020×20×2010×10×20

Artifact Along Block Boundaries 20×20×2010×10×20

Summary Importance-driven data analysis and visualization Quantify data importance using conditional entropy Cluster the importance curves Leverage the importance in visualization Limitations Block-based classification Size of joint feature-temporal histogram Extensions Non-uniform data partition Incorporate domain knowledge Dimension reduction

Acknowledgements NSF CCF , CNS , OCI , OCI , and OCI DOE SciDAC Program DE-FC02-06ER25777, DE-FG02-08ER54956, and DE-FG02- 05ER54817 Data sets Combustion: Jacqueline H. Chen, SNL Climate: Andrew T. Wittenberg, NOAA Earthquake: CMU quake group Hurricane: NSF, IEEE Visualization 2004 Contest