Charts are better than nothing, but you really want to have the numbers in Excel or Matlab so you can do analysis and get big insights (or at least make a nicer-looking chart).

The answer is Plot Digitizer

Below is a short tutorial on using this tool to go from chart to data faster and more accurately than simply “eye-balling” it.

Using Plot Digitizer

Step 1. Download Plot Digitizer. Then unzip the files to someplace safe (e.g., Program Files, Applications, etc.).

Step 2. Get the chart. Save a .png file of the graph you’d like to digitize (.png files work best because they use loss-less compression and Plot Digitizer can read them). A screenshot is usually the easiest way to grab the chart. (simple tutorial on screenshots). Also, remember to zoom in before taking the screenshot so it’s nice and big.

Step 3. Start up. Run PlotDigitzer (just double-click the .exe) and open the .png image.

Step 4. Calibrate the axes. You’ll jump right into calibration when you open the image. The cursor becomes a cross-hairs, and instructions appear at the bottom of the screen. Start by clicking the far left side of the x axis, and entering the value when prompted. Remember to click where the data is defined (e.g., click in the center of first bar in the example below).

Continue calibrating by clicking the far-right end of the x axis, both extremes of the y axis, and labeling both axes.

Step 5. Digitize. Click on each datapoint in the chart that you’d like to capture. You can do this multiple times if the chart has multiple data series (e.g., in a multi-line line chart or a bar chart with split bars). Try to click as close as possible to the true spot on the chart. If you miss a point you can always undo clicks with the button at the top.

When you finish, click “Done” at the far right of the second toolbar.

Step 6. Export the data. Either select the cells in the output windows and paste the data into Excel, or go to “File->Save As” to save a .csv file.

Use that data

And there you have it! You now have (almost) the exact data used to generate the chart. If you’ve got any other tricks to extract data from onerous sources, let us know in the comments!

• Part 1: Branding Yourself and Making Making a “Wow” Resume
• Part 2: Preparing for Your Case Interview to Get Bulletproof
• Part 3: Talking about Your Experience and Sounding like a Bad-ass

Leaving academia and joining consulting firms is a something many PhD students (myself included) are getting interested in. Firms like McKinsey & Company, Boston Consulting Group (BCG), and Bain & Associates once hired mostly MBAs but are now branching out to hire MDs, lawyers, and PhDs.

I wanted to make a big impact with the skill I learned during my PhD. I got excited when I heard about consulting because it promises just that. In the next three parts, I’ll take you through the big lessons I learned while preparing and interviewing: How to get an interview, how to nail the case, and how to dazzle them with your experience.

Part 1: Branding Yourself and Making Making a “Wow” Resume gives you pointers to polish that scruffy science look off your C.V. and generally control your “personal brand” so that interviewers are impressed with you long before you walk in the door.

Part 2: Preparing for Your Case Interview to Get Bulletproof talks about how to approach the case and how to practice so that you can shine while others look dull. I’ll give you some simple exercises that will improve the structure and creativity of the “case” portion of your interview.

Part 3: Talking about Your Experience and Sounding like a Bad-ass covers an important and often overlooked portion of a consulting interview… talking about yourself! I know you have some amazing stories to tell. This sections shows how to make your stories say the right things about you.

Please enjoy!

Disclaimer: I recently went through the application and interview process with a top firm, came out with an offer, and signed it! In this series, I share my experience and give some ideas for people on a similar path. However, at the time of writing (July 2009), I do not have any inside information on how any company conducts their hiring. These are just my thoughts!

First I’ll talk about how to build your brand, then I’ll share my tips on resume writing.

Building Your Personal Brand

The first thing I do when I hear about a person is google them to see what they’ve done. The google search results for someone’s name tells a story, and you have the power to control your own story!

To shape the way new people (specifically interviewers) see you on the web you’ll need to think strategically about the story you want to tell and then go get famous for things you’re good at.

Thinking Strategically:

Think about the person that you would like to be and the person that the interviewers are dying to hire. Find where those two people overlap and start thinking strategically about how you can become that person.

It is best to get started early, but it is worth considering at any time. The biggest rewards come from lots of hard work, but you can also make big improvements by being strategic about presenting what you’ve already done.

Let’s say that your goal is to become an, “innovator with a keen business sense and powerful communication skills.” Well, let’s break that down:

• Innovator → Show examples of things you created.
• Business Sense → Have stories of business you’ve been involved in.
• Communication Skills → Communicate well in lots of different mediums.

Read company websites. They’ll often tell you exactly what the company is looking for. If there’s a particular job or firm that you really like then find out what they want and include those things as part of your personal brand.

Find some examples. If you already have great examples of things you’ve created, business scenarios you thrived in, and your communication abilities, you might be done. Otherwise, come up with a plan to fill out those areas.

Get new experiences if you don’t already have them. To become an “innovator,” find where you can add some value by creating something new and get to it. To get “business sense,” take some classes or get involved a venture where you can gain experience and help out at the same time. To show what a great “communicator” you are, you might start a blog. Well… that leads us to getting famous.

Getting famous:

This gets us back to the google results for your name. The goal here is to have your google results page read like a resume. You can make this happen by:

• Getting member pages on web sites for research and professional groups
• Taking leading rolls in events and conferences with web presences
• Contributing to scientific journals with high impact factors
• Creating your own content and sharing it on a personal web-page

Tell a very compelling story about yourself, your accomplishments, and your worth to an employer by creating a blog or personal web-page filled with content that describes your research, solves problems for others, and describes other aspects of your professional life. That’s what I’ve done here, if you haven’t noticed : )

Clean up your facebook page and/or make it private. That should really go without saying, but… yeah. Also fill out your linkedin profile with relevant info and get some recommendations.

Once you have the right experience, and your google results tell your story for you, you’ll have a much easier job putting together a winning resume.

Making a “Wow” Resume

Clearly communicate your brand with your resume. Most of the resumes I see are hard to read, ugly to look at, and filled with jargon. This is sometimes acceptable for a scientific CV, but if you’re stepping out of academia it is time to step up the resume.

Here are some major things to consider:

• Give ‘Em the Business. The types of projects that PhDs have on their resumes can be very specific and jargon-y. Re-write your experience to talk about the higher-level objectives and accomplishments.
• Include the “So What.” If the reader doesn’t know your field, they may not care about the specifics of your projects. What most people can understand is the “so what.” After a SHORT description of a technical accomplishment, specify the specific result. “This work resulted a patent application,” or “This work resulted in 4 publications.”
• Be Specific. You may think being vague about past projects makes them seem more impressive. It doesn’t. People often assume the worst. By sticking to the facts and being specific you’ll appear honest and accomplished.
• Include numbers. Use numerals to add even more specificity and help readers find facts. “Led a group of 10 people.” or “Organized a conference with 30 participants.”
• Use Bolding and Bullets. Just like writing for the web, it is important to make sure a resume is skim-able. If there are large blocks of simple text you can assume people won’t read them.
• Typography and Whitespace. Make sure the typography looks attractive and has plenty of white space around it. Don’t be afraid to get a little artsy with the fonts, but don’t go crazy.
• Mind Page Breaks. It is acceptable to have a 2-page resume, but make sure those page breaks fall at logical places. Don’t cut a section or sentence in the middle of a page.
• Edit, Edit, Edit. Proof reading is a must, but revising your resume again and again (and again) will make the descriptions of things you’ve done clear, concise, and powerful. LOTS of editing transforms resumes from rambling messes to powerful marketing documents.

After you have drafted your resume, you should have it checked by as many different people as you can and take their criticisms to heart. The same goes for cover letters and application essays. Often times you will not recognize what a third party gets out of your writing until they tell you. Don’t let the only “third person” be the interviewer.

Leave your ideas about branding and resumes in the comments.

Next: Case Interview

Now that you’re famous with an awesome resume, check out:
Part 2: Preparing for Your Case Interview to Get Bulletproof, where I talk about how to approach a case interview and how to practice so that you can shine while others look dull. I also share some simple exercises that can improve the structure and creativity of the “case” portion of your interview.

• Series Intro: A PhD’s Guide to Getting a Consulting Job
• Part 1: Branding Yourself and Making Making a “Wow” Resume
• Part 2: Preparing for Your Case Interview to Get Bulletproof
• Part 3: Talking about Your Experience and Sounding like a Bad-ass

Case interviews are an interview tool that consulting companies use to gauge your analytical skills. Essentially, the case interview involves answering an open-ended question about a business problem. The interviewer gives some background on a (hypothetical) company that needs help. Then the candidate talks through an analysis and solution.

I’m going to show you how to rock a case interview.

The interviewers are looking for a few things in particular in a case interview:

• Structured thinking
• Hypothesis-driven approach
• Strong analytical ability
• Creative problem-solving skills
• Keen business sense
• Composure in the face of pressure and uncertainty

McKinsey-style Cases Vs. Normal Cases:

Some firms use “McKinsey-style” cases to make the interview more fair and relieve some pressure. These cases are more structured and give the applicant three or four specific questions to answer. “Normal” cases, on the other hand, give a problem statement and expect the applicant to direct the entire case with little or no guidance from the interviewer. It is a good idea to practice both, as they each teach different skills.

You can find good resources and sample cases at the Georgia Tech Consulting Club site.

Things to Remember

If you read some of the material available, you’ll get lots of advice. These are the 5 things that I think are most important.

• Structure Everything. EVERYTHING. Instead of jumping right into an answer, start by outlining your answer to the interviewer. First, describe high-level “buckets.” Next, discuss each one in turn by listing lower-level items within each “bucket.” Think trees. You should give an overall framework for solving the case and separate frameworks for estimation problems, math problems, data analysis, final recommendations… EVERYTHING. The more you structure, the better you’ll look.
• State Your Hypothesis. Never ask the interviewer for information or start making analysis without specifying what you’re attempting to prove (or disprove). You can demonstrate that you are “hypothesis driven” by stating your hypotheses out loud and saying how you plan to evaluate them. This sentence would be a good example, “I would guess that competition is a key driver of the low price because widgets are a commodity. Can you share any information on the competitive landscape to help me confirm that?”
• Back up Assumptions. You’ll often be asked to make assumptions. Partly this is to see if you make reasonable guesses. However, always make educated guesses and then specify how you could go about collecting the real information if needed. For example, “I’d assume 25% of personal automobiles are trucks based on what I see driving around. Of course, I could check that by looking at industry sales numbers. Does 25% sound like a reasonable estimate for now?”
• Ask “So What?” Every time you reach a number, conclusion, or stopping place immediately ask yourself, “So what?” Then answer out loud to the interviewer. Don’t say, “It will cost $2 million.” Instead, say, “It will cost $2 million, which is [surprising, disappointing, expected, etc.] considering…” By stepping back and answering “So what?” without being prompted you show that you can grasp the big-picture. And you’ll make an impression.
• Nail the recommendation. The last thing you’ll do in a case is “give your final recommendation.” Nail it by being direct and decisive. Have the first words out of your mouth be what the company should do. Cite a couple of reasons why, give one or two caviots, re-state the recommendation and stop. Whatever you do, don’t summarize the case to the interviewer… they were there! They know!

How to Practice

You can build up your case interview skills before you ever practice a case with someone else. However, you should definitely get some mock-interviews under your belt before your real interview. Here are 3 ways to practice. 2 you can do solo, the other you need a partner for.

1. Structure Drills

This drill will help you get your “initial structure” handled. This crucial step can make or break a case, and if you do it right, you will be off to a great start. The secret is doing this enough that it becomes second nature.

Read the case intro. Start by finding a case (try looking here). Read the intro paragraph describing the case, but nothing else. Don’t look at any additional information yet.

Work for 10 minutes. Brainstorm and write a detailed tree-shaped structure for approaching the case. Your final tree might look something like this. However, yours should be more detailed, and tailored to the specific case you’re working with.

Keep the structure MECE (mutually exclusive and completely exhaustive). This means that each category is independent from the others, but together they cover all possible areas needed to analyze the case. Edit your structure until you feel that it is as good as it can get.

Improve your structure. Read the rest of the example case and re-evaluate your work. Did you leave anything out? Would a different structure have made more sense in light of the new information? Edit and improve your structure-tree again. Note any revelations that may help you next time.

Practice presenting it in 30 seconds. In the interview, you’ll have about a minute to prepare a structure and then 30 seconds to present it to the interviewer. You want to start with a bang, so make sure you practice presenting it OUT LOUD. Start by saying the top-level items and then going into specifics for each. Make sure you convey which top-level item you hypothesize is the most important and why.

Do this drill over and over until you are a case-structuring-machine!

2. Creativity Drills

Interviewers sometimes ask for you to come up with “some ways to cut costs,” or “some new ways to market the product.” These questions are meant to test for creativity and out-of-the-box thinking. To some extent, this has to do with how your brain works. But it you can still warm up and get in the habit of thinking creatively.

As you go through your day, come up with creativity-prompting questions. If you stop for coffee ask yourself, “how could they get me to pay $0.25 more per cup.” When you get groceries ask yourself, “what are some ways this grocery store could cut costs.” By doing this all the time you’ll get in the habit.

Make lists of 10, 20, 50 things. One way to coax out creativity is to number a page (or open a spreadsheet) and commit to making a list of 10, 20, or 50 creative answers to some of the questions you came up with during the day. Don’t worry about “good” or “bad” ideas, just make sure you get to your number-goal. You’ll be surprised how inspiring that big empty list can be.

3. Mock Cases

Practicing by yourself is necessary to hone certain skills, but nothing compares to sitting across the table from a friend to do some simulated interviewing. Here are a few tips on making the best of practices.

Find good people to help you. The best choices are people who are currently working at the firm you’re applying to. The next best choices are people who have been through these types of interviews. Be careful about how you use these peoples’ time. Don’t use your best resources until your skills are sharp from practicing alone and with less experienced friends.

Be serious throughout the entire mock interview. Don’t joke around just because it’s your friend and it’s “only practice.” Ask your partner not to pull any punches. The stress helps you learn. You want a practice that is as close as possible to the real thing.

Share feedback right after the interview. The interviewee should start off by giving their impression of the case. Even if you’re case skills need work, it is important to make sure your “interviewer” sees the same problem areas as you do. When the “interviewer” gives feedback, take notes and ask questions. This is probably the best information you can get on how to improve.

Leave questions or other preparation ideas in the comments.

Next: Experience Interview

Now you know how to practice for your case interview. Don’t forget:
Part 3: Talking about Your Experience and Sounding like a Bad-ass, which covers an important and often overlooked portion of the interview… talking about yourself! I know you have some amazing stories to tell; learn how to make them say the right things about you.

• Series Intro: A PhD’s Guide to Getting a Consulting Job
• Part 1: Branding Yourself and Making Making a “Wow” Resume
• Part 2: Preparing for Your Case Interview to Get Bulletproof
• Part 3: Talking about Your Experience and Sounding like a Bad-ass

Disclaimer: I recently went through the application and interview process with a top firm, came out with an offer, and signed it! In this series, I share my experience and give some ideas for people on a similar path. However, at the time of writing (July 2009), I do not have any inside information on how any company conducts their hiring. These are just my thoughts!

Experience questions break into two types; simple chit-chat questions and serious in-depth questions. These are both quite important, and I’ll talk about how to prepare for each:

Preparing for Chit-Chat Interview Questions

You should probably have 30-90 sec. answers to a variety of likely “chit-chat” questions in mind.

• “So, why are you interested in consulting?”
• “What is the lay-person version of what your thesis is about?”
• “Why abandon your research when it seems so important?”
• “I won’t ask if your interviewing with our competitors, but… why our firm?”
• “How was your morning?”

Have your answers ready. These questions might get asked in the very beginning of the interview before you even know that it has started. This chit-chat can be very important, though. You’ll want to show that you have thought carefully about why you are interested in consulting and why the firm you’re interviewing with is a good fit.

Preparing great answers to these questions can also help you explain what you’re doing to your friends and family!

Example In-Depth Experience Questions

These questions should give you an idea of the kinds of things you’ll be asked. You can make up hundreds more questions by varying the themes in the samples below.

• Give me an example of a time that you had to take a team in a new direction in order to save a project under a tight deadline.
• Tell me about a time when you set a specific goal for yourself and accomplished it.
• What was the biggest challenge within your PhD and why?
• Give me an example of a time when you tried to change someone’s mind in an important situation. (Could be successful or unsuccessful) For this example, try to think of something where the person was very firm about their position?
• Consider a time when you were forced to go against a group decision. How did you know that you had to do what you did, and what did you do afterwards?
• Tell me about a time when you had to tell a superior or supervisor they were wrong.
• Tell me about a time when you worked with someone that was very challenging. Try to think of a situation in which you had to directly confront this person.
• Recall a time when you had to confront someone for poor work on a project.
• Tell me about a time when you had to influence a large group of people (try to think of something that isn’t school/work related).
• Give me an example of when you had to lead a team through a crisis?
• Many times in consulting we have to work directly with CEOs and other executives. Tell me about an experience when you had to meet with someone very high up in an organization.

Preparing for In-Depth Experience Questions

In-depth questions like those above help the interviewer gauge how you think and how you have solved real problems in your past. They also will cause you to show a lot of your personality if you answer them well. Here are some things to remember.

Prepare at least 5 stories. Each story should be flexible and detailed enough to answer two or three of the sample questions depending on how you tell it. Having at least 5 stories in mind should give you plenty of variety to handle any question the interviewer throws at you.

Use a compelling title. If you have a story about how you had to convince your boss to re-write a chapter in a grant application (for example), Don’t just jump into the story. First, present the title: “This story is about a time when I influenced my supervisor to make a last-minute change that saved our group’s funding.” The short, exciting synopsis sets the tone for the rest of the question.

Have at least 10 minutes of talk for each story. Hopefully you’ll get some breaks, and you won’t have to talk the whole time. Fill your stories with hooks that deliberately bait the interviewer into asking further questions (which you also have answers to). Be interactive throughout the story. Ask the interviewer if they’re following and stop to answer any tangential questions the interviewer may have.

Don’t forget your structure. Prepare by remembering all the details, writing down key bits of information, and trying to impose a structure on the narrative. Come up with lots of 3-bullet-lists to use in the story. 3 Observations you had, 3 problems you faced, 3 things you tried, etc. These will lend structure to the story, help the interviewer know where you’re headed, and keep you focused.

Consider a layout like this:

1. Start with a hook or “headline” (10 sec.)
2. Set up by telling why its important and how it relates to the question (~30-60 sec.)
3. Mention what the 3 major problems were. (10 sec.)
4. Talk about the 1st problem and the solution (120 sec.)
5. Talk about a 2nd problem and the solution (120 sec.)
6. Talk about a 3rd problem and the solution (120 sec.)
7. Give a nice conclusion and epilogue and say what you learned. (30-60 sec.)

Be very detailed about your thought process. Shoot to give details without being prompted, but realize that the interviewer might jump in and ask for clarifications here and there. A big difference between consulting interviews and normal interviews is that consultants want the details. Be sure to tell why you did things, how you reached your conclusions, and what specifically you did about them.

It’s like being at a party. You want to be engaging, suspenseful, and impressive. You should let people jump in to ask questions, but you should be able to roll with the story if no one does. This also means you should use real stories. Pull from research, class projects, past jobs, personal life, hobbies, etc., but don’t make stuff up.

Other ideas for practicing for experience interviews? Leave a comment.

Conclusion

Well, folks. That wraps it up. We’ve covered building your brand, honing your resume, prepping for case interviews, and now, talking about your experience. With these tools and some careful practice you should be able to go in and rock those interviews!

I wish you the best of luck in finding the consulting job of your dreams!

• Series Intro: A PhD’s Guide to Getting a Consulting Job
• Part 1: Branding Yourself and Making Making a “Wow” Resume
• Part 2: Preparing for Your Case Interview to Get Bulletproof
• Part 3: Talking about Your Experience and Sounding like a Bad-ass

Disclaimer: I recently went through the application and interview process with a top firm, came out with an offer, and signed it! In this series, I share my experience and give some ideas for people on a similar path. However, at the time of writing (July 2009), I do not have any inside information on how any company conducts their hiring. These are just my thoughts!

It took a bit of searching, but eventually I found RSS feeds for many of the journals I’m interested in and loaded them into google reader. It is now quick to scroll through new abstracts as papers appear on-line prior to publication. Below are links to RSS feeds for some computer vision journals I’m keeping up with.

RSS Feeds for Computer Vision Journals

Finding RSS Feeds for Other Journals

It takes a bit of hunting sometimes, but I can’t imagine that a journal would not have RSS these days. IEEE Journals are easy to find, and I found that inezha.com was a good resource for finding some of the other ones I have listed.

Any good feeds I missed?
Other good ideas for keeping current?

Leave them in the comments.

Use Fast Level-Sets

Start by using a fast level-sets implementation that minimizes the number of required computations [code]. This will already save a huge number of computations per iteration and speed things up quite a bit!

Create better initializations.

The farther the initial contour is from its final position, the more computations must be done for the contour to converge. Hence, if you can start the contour in almost the right place, you’ll drastically reduce the time needed for segmentation. You can use prior knowledge, user input, or other segmentation techniques to create a rough guess that is close to the right answer. Another initialization that can leads to quick initialization is ‘bubbles’ on an evenly-spaced grid.

Use a multi-scale approach.

This is a way to quickly get good initializations using active contours. Say your data is MxN. Instead of segmenting the full data set, downsample the data so that you are dealing with an (M/8)x(N/8) volume. The segmentation should run much quicker on the smaller volume. Next, upsample the result back to MxN and use this as an initialization for the full data. The idea is that the time saved on the full segmentation by having a good estimate based on downsampled data will make up for the time needed to downsample, segment on the small data, and upsample.

Use approximate active contours.

Using an approximate solution for all or part of your segmentation can be helpful. As in 2 and 3, you can use an approximate active contour technique to quickly get close to the right answer. Then you can use an accurate level sets implementation to get the right answer quickly. Alternatively, the discrete methods can work quite well alone! James Malcolm proposed a nice method in “Fast Approximate Surface Evolution in Arbitrary Dimension” [code].

Use another technique entirely.

Active contours are “variational,” so they give nice, principled solutions with analytic geometry, etc. However, if you just want fast segmentations, other techniques such as thresholding/morphology, graph cuts, region growing, etc. can all be viable solutions.

Any other tips or links to good implementations? Leave them in the comments.

Fast Level Sets Demo

The links below point to the technical report and a demo written in C++/MEX that can be run directly in MATLAB. The demo implements the Chan-Vese segmentation energy, but many energies can be minimized using the provided framework.

Sparse Field Method – Technical Report [pdf]
Sparse Field Method – Matlab Demo [zip]

To run the MATLAB demo, simply unzip the file and run:
>>sfm_chanvese_demo
at the command line. On the first run, this will compile the MEX code on your machine and then run the demo. If the MEX compile fails, please check your MEX setup. The demo is for a 2D image, but the codes work for 3D images as well.

My hope is that other researchers wishing to quickly implement Whitaker’s method can use this information to easily understand the intricacies of the algorithm which, in my opinion, were not presented clearly in Whitaker’s original paper. Personally, these codes have SUBSTANTIALLY sped up my segmentations, and are allowing me to make much faster progress towards completing my PhD!

Thanks to Ernst Schwartz and Andy for helping to find small bugs in the codes and documentation. (they’re fixed now!)

For more information regarding active contour, segmentation, and computer vision, check here: Computer Vision Posts

On a side note, people interested in creating screen-casts of presentations on a Mac, should consider the program ScreenFlow, which worked great for me! This was also my first presentation created with Apple’s Keynote software, but I’m sure it won’t be the last.

This post will cover some tips, tricks, and scripts to get your journal submission ready to go quickly and easily. The big challenges are:

• Getting all figures into single eps files
• Creating a list of figure captions

The good news is that both of these are delightfully quick if you’ve been using LaTeX to compose your article. Read on to find out how

Get all figures into .eps files

For some figures, it may be simple to use convert or any other image editing software to convert simple figures into eps files. With convert, simply type this at a linux or mac terminal:

$ convert image_from_paper.png fig1.eps

If you’ve used the subfigure package, PSTricks, tables, or other funny-business for creating or formatting figures within LaTeX, this process isn’t as straightforward. But by following these simple steps you can get it done quickly.

Create a file: figs_only.tex

figs_only.tex is a file that just contains the code to make and format the figures. Begin by formatting the paper into the two-column format as best you can. Then, copy TeX codes that make the figures into this file:

figs_only.tex

Follow the pattern of the figures in the file and don’t copy over any caption information, just the TeX code to format and place the figures. Also, remember to un-comment the appropriate parts of in the header if you are using PSTricks or subfigure packages.

Note the use of the \clearpage command after each figure environment. This makes a new page for each figure. Also, note the \pagestyle{empty} command at the top. This removes any page numbers so that the figure is the only thing on each page. Finally, when you use subfigure, you’ll have to reset the counter manually by using \setcounter{subfigure}{0} before each figure that uses subfigures.

Get to figs_only.pdf

Great! Now, run LaTeX to create figs_only.dvi. You can follow this up with a call to dvipdf to create figs_only.pdf.

$ latex figs_only
$ dvipdf figs_only

I’ve heard other suggestions to use dvips and then ps2pdf instead of dvipdf, but dvipdf has always worked fine for me. Also, if you aren’t using PSTricks, you can use pdflatex and skip the whole thing! (PSTricks doesn’t play nice with pdflatex, though).

Convert the pdf file to eps figures

This is the really great part. Now that you have a pdf file with each of your figures on a separate page, you can run pdf2eps (from this post). “pdf2eps ” builds an eps file from the page specified. It uses pdfcrop (from here) to trim all whitespace. Hence, all you need to do is run

$ ./pdf2eps.sh 1 figs_only
$mv figs_only.eps fig1.eps
$ ./pdf2eps.sh 2 figs_only
$mv figs_only.eps fig2.eps
… etc.

Voila! now you have eps files for each figure in your article. Remember to go back every figure in the original paper to use the eps file rather than whatever complex LaTeX codes you used to format the figure before.

pdf2eps.sh
pdfcrop.pl

Create the List of Figures

This part is lots simpler than the first! If you’ve used \caption{} within your figure environments, there is a command that will automatically build your list of figures and captions for you. For best results, put this at the end of the .tex file right before \end{document}. This ensures that page numbers are correct.

\clearpage
\setcounter{page}{1}
\listoffigures

Now, when you re-build your article there will be a list of figures nicely formatted and ready to go. You can either print these pages separately or use Adobe to split of the last pages of the document as a separate pdf file. (Does anyone know how to do this without the need to split the document?)

Submit that puppy!

These tips help you prepare the electronic submission to the IEEE journals. Remember to follow all the little rules that may be specific to each publication. Also, please leave some comments if you have any other tips to get these submissions put together.

Finally, special thanks to tug.org for this post which I drew from extensively.