What is the main drawback of the minimization of informational divergence?

The main drawback of the minimization of informational divergence is that the final transfer function could be given by too low opacity values, i.e., the final solution could be too transparent.

What are the basic information-theoretic measures used in computer graphics?

In computer graphics, the most basic information-theoretic measures have been used in scene complexity [8], global illumination [27], light positioning [11], and viewpoint selection for polygonal scenes [35, 32, 9].

How does the volume data set be rendered?

Once the program has converged, or has done all the iterations, the volume data set is rendered using the optimal transfer function.

What is the way to get the intensity values of the voxels inside the mask?

If the authors assume that the original volume data set takes intensity values from 0 to max, a new data set is generated by keeping the intensity values of the voxels outside the mask and adding the value max+1 to the intensity values of the voxels inside the mask.

How is the transfer function specification done?

Transfer function specification with this approach is accomplished by selecting previews from a randomized selection to guide the search process.

What is the meaning of the definition of informational divergence?

As the authors have mentioned in the previous section, the informational divergence can be interpreted as a distance between the observed data and a theoretical description which can include the relevance of data.

Why is the informational divergence based on absolute visibility?

This is due to the fact that the informational divergence is based on probability distributions, which can be seen as ratios of visibility, and not on absolute visibility values.

Why does the method assign high opacities to the bone structure?

Observe that when weighting occurrences by intensity (Figure 5(e)) the method assigns high opacities to the bone structure because this has the highest intensity values in CT images.

What is the common measure of probability distribution distance in information theory?

A common measure of probability distribution distance in information theory is the informational divergence or Kullback-Leibler distance [7, 45] (see Equation 6).

What is the way to tackle the problem?

In this case, an analytical computation, or at least a reasonable good approximation, becomes the best way to tackle the problem.

How many iterations are required to compute the CT-carp?

For each configuration the authors collect the computation time in seconds and the number of iterations required by different target distributions.

Open AccessJournal Article10.1109/TVCG.2011.173

Automatic Transfer Functions Based on Informational Divergence

M. Ruiz, +5 more

- 01 Dec 2011

- IEEE Transactions on Visualization and C...

- Vol. 17, Iss: 12, pp 1932-1941

TL;DR: This paper presents a framework to define transfer functions from a target distribution provided by the user, based on a communication channel between a set of viewpoints and aSet of bins of a volume data set, and it supports 1D as well as 2D transfer functions including the gradient information.

Abstract: In this paper we present a framework to define transfer functions from a target distribution provided by the user. A target distribution can reflect the data importance, or highly relevant data value interval, or spatial segmentation. Our approach is based on a communication channel between a set of viewpoints and a set of bins of a volume data set, and it supports 1D as well as 2D transfer functions including the gradient information. The transfer functions are obtained by minimizing the informational divergence or Kullback-Leibler distance between the visibility distribution captured by the viewpoints and a target distribution selected by the user. The use of the derivative of the informational divergence allows for a fast optimization process. Different target distributions for 1D and 2D transfer functions are analyzed together with importance-driven and view-based techniques.

AI Agents for this Paper

Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps

Most frequently asked questions

1. What have the authors contributed in "Automatic transfer functions based on informational divergence" ?

In this paper the authors present a framework to define transfer functions from a target distribution provided by the user.. Their approach is based on a communication channel between a set of viewpoints and a set of bins of a volume data set, and it supports 1D as well as 2D transfer functions including the gradient information.

2. What future works have the authors mentioned in the paper "Automatic transfer functions based on informational divergence" ?

Future research will be done to define a new objective function that approximates the target distribution while maximizing the projected visibility without requiring a parameter such as β.. Future work will be done to obtain a fully GPU implementation.

3. What is the main factor that affects the visibility of structures?

In their approach, the authors have focused on opacity transfer functions considering that opacity is the main factor that affects the visibility of structures.

4. How many uniformly distributed viewpoints have been used in the following experiments?

In the following experiments, the authors have used by default the global informational divergence (GID), β = 0, a stopping threshold value of the objective function equal to 0.001, and 6 uniformly distributed viewpoints.

Table 1. Number of iterations and time cost (in seconds) for different data sets and stopping threshold values. Target distributions are: occurrence (Occ), occurrence weighted by importance (Occ*Imp), occurrence weighted by gradient and intensity (Occ*Grad*Int), and occurrence weighted by gradient (Occ*Grad).

Fig. 4. Evolution of the objective function F(A) for GID with respect to the number of iterations for the CT-body using the occurrence target distribution, β = 0, and 6 viewpoints.

Fig. 5. Global strategy applied to the CT-head. The following target distributions have been considered: 1D transfer functions with occurrence (Occ) and occurrence weighted by intensity (Occ*Int); 2D transfer functions with occurrence weighted by gradient (Occ*Grad) and occurrence weighted by both gradient and depth (Occ*Grad*Depth). The global informational divergence (GID) has been used in (a, b, e, f), the Euclidean distance in (c, d), and the viewpoint informational divergence (VID) in (g, h), where only the viewpoint shown has been used.

Fig. 8. Visualization of the CT-body obtained from the target function given by the occurrence weighted by gradient and β values equal to 0, 2/3, and 8/9, respectively.

Fig. 6. Importance-based strategy applied to the CT-head assigning importance 1 to the intensities corresponding to bone structure and 0.2 to the rest, using as target distributions: (a) occurrence and (b) occurrence weighted by gradient.

Fig. 7. Importance-based strategy applied to two MRI brain data sets: (a) initial visualization of the MRI-healthy and the resulting visualization giving importance to (b) white matter, (c) grey matter, (d) glial tissue. (e) Visualization of the MRI-damaged giving importance to the lesion area. In (b) and (c) the importance is 1 for the masks and 0.2 for the rest. In (d) and (e) the importance is 1 for the masks and 0.1 for the rest.

Citations

Journal Article•10.1016/J.MEDIA.2017.01.007

The status of augmented reality in laparoscopic surgery as of 2016.

Sylvain Bernhardt, +3 more

- 01 Apr 2017

- Medical Image Analysis

TL;DR: A comprehensive review of all the different methods proposed by the literature concerning augmented reality in intra-abdominal minimally invasive surgery (also known as laparoscopic surgery) in order to better grasp the current landscape of the field.

...read moreread less

296

•Journal Article•10.1109/TVCG.2018.2816059

A Generative Model for Volume Rendering

Matthew Berger, +2 more

- 01 Apr 2019

- IEEE Transactions on Visualization and C...

TL;DR: This work uses the Generative Adversarial Network framework to compute a model from a large collection of volume renderings, conditioned on viewpoint and transfer functions for opacity and color, and transforms transfer functions into a view-invariant latent space specifically designed to synthesize volume-rendered images.

...read moreread less

•Journal Article•10.1109/TVCG.2012.195

An Affordance-Based Framework for Human Computation and Human-Computer Collaboration

R. J. Crouser, +1 more

- 01 Dec 2012

- IEEE Transactions on Visualization and C...

TL;DR: This work has reviewed 1,271 papers from many of the top-ranking conferences in visual analytics, human-computer interaction, and visualization, and identified 49 papers that are representative of the study of human- computer collaborative problem-solving, and provides a thorough overview of the current state-of-the-art.

...read moreread less

•Journal Article•10.1016/J.MEDIA.2016.01.008

Automatic localization of endoscope in intraoperative CT image: A simple approach to augmented reality guidance in laparoscopic surgery

Sylvain Bernhardt, +5 more

- 01 May 2016

- Medical Image Analysis

TL;DR: This article proposes a new approach to automatically register the reconstruction from an intraoperative CT acquisition with the static endoscopic view, by locating the endoscope tip in the volume data using standard image processing algorithms.

...read moreread less

•Journal Article•10.3390/E20050370

A Survey of Viewpoint Selection Methods for Polygonal Models.

Xavier Bonaventura, +4 more

- 16 May 2018

- Entropy

TL;DR: This survey reviews and compares twenty-two measures to select good views of a polygonal 3D model, classify them using an extension of the categories defined by Secord et al., and evaluates them against the Dutagaci et al. benchmark.

...read moreread less

...

Expand

References

Journal Article•10.1002/J.1538-7305.1948.TB01338.X

A mathematical theory of communication

Claude E. Shannon

- 01 Jul 1948

- Bell System Technical Journal

TL;DR: This final installment of the paper considers the case where the signals or the messages or both are continuously variable, in contrast with the discrete nature assumed until now.

...read moreread less

74.4K

•Book

Elements of information theory

Thomas M. Cover, +1 more

- 01 Jan 1991

TL;DR: The author examines the role of entropy, inequality, and randomness in the design of codes and the construction of codes in the rapidly changing environment.

...read moreread less

52.2K

•Journal Article

The mathematical theory of communication

Claude E. Shannon, +1 more

- 01 Jan 1949

- IEEE Transactions on Instrumentation and...

TL;DR: The Mathematical Theory of Communication (MTOC) as discussed by the authors was originally published as a paper on communication theory more than fifty years ago and has since gone through four hardcover and sixteen paperback printings.

...read moreread less

36.2K

Numerical recipes in C

William H. Press, +3 more

- 01 Jan 1994

TL;DR: The Diskette v 2.06, 3.5''[1.44M] for IBM PC, PS/2 and compatibles [DOS] Reference Record created on 2004-09-07, modified on 2016-08-08.

...read moreread less

20.7K

•Journal Article•10.1109/38.511

Display of surfaces from volume data

Marc Levoy

- 01 May 1988

- IEEE Computer Graphics and Applications

TL;DR: In this article, a volume-rendering technique for the display of surfaces from sampled scalar functions of 3D spatial dimensions is discussed, which is not necessary to fit geometric primitives to the sampled data; images are formed by directly shading each sample and projecting it onto the picture plane.

...read moreread less

2.5K

...

Expand

Automatic Transfer Functions Based on Informational Divergence

Chat with Paper

AI Agents for this Paper

Most frequently asked questions

1. What have the authors contributed in "Automatic transfer functions based on informational divergence" ?

2. What future works have the authors mentioned in the paper "Automatic transfer functions based on informational divergence" ?

3. What is the main factor that affects the visibility of structures?

4. How many uniformly distributed viewpoints have been used in the following experiments?

5. What is the main drawback of the minimization of informational divergence?

6. What are the basic information-theoretic measures used in computer graphics?

7. How does the volume data set be rendered?

8. What is the way to get the intensity values of the voxels inside the mask?

9. How is the transfer function specification done?

10. What is the meaning of the definition of informational divergence?

11. Why is the informational divergence based on absolute visibility?

12. Why does the method assign high opacities to the bone structure?

13. What is the common measure of probability distribution distance in information theory?

14. What is the way to tackle the problem?

15. How many iterations are required to compute the CT-carp?

Figures

Citations

The status of augmented reality in laparoscopic surgery as of 2016.

A Generative Model for Volume Rendering

An Affordance-Based Framework for Human Computation and Human-Computer Collaboration

Automatic localization of endoscope in intraoperative CT image: A simple approach to augmented reality guidance in laparoscopic surgery

A Survey of Viewpoint Selection Methods for Polygonal Models.

References

A mathematical theory of communication

Elements of information theory

The mathematical theory of communication

Numerical recipes in C

Display of surfaces from volume data

Related Papers (5)

Visibility Histograms and Visibility-Driven Transfer Functions

Semi-automatic generation of transfer functions for direct volume rendering

Size-based Transfer Functions: A New Volume Exploration Technique

Multidimensional transfer functions for interactive volume rendering

Curvature-based transfer functions for direct volume rendering: methods and applications