The photo above is a mockup. The actual tool creates annotations with much thinner lines.

The first ImageStats tasks will be posted Monday, November 25th. A slow-rollout/pilot-phase (starting with 3 tasks) is anticipated. This description is subject to change during the pilot phase.
This page last updated Nov 25st.

If you participated last year, you will recall whole-face and skin-patch annotation tasks. This year, there is only one type of task, different from either of those, which is locate and annotate the eyes, nose, and mouth, collectively called “landmarks”. Last year's whole-face task required differentiating between front-face and side-face poses; not so this year. Also, there are no pre-computed rectangles to adjust; you start with an unannotated photo in every case. Each task is composed of more photos this year (600 instead of 400). Also, this year, you have to get from us a different username & password for every task.

Familiarity with using a web browser. Also, the need to judge certain rectangle colors, which may impede those with certain forms of color blindness. No software coding is involved.

The task requires access to a recent browser that has HTML 5 support. Chrome is recommended.

Research is going on to develop and improve face-detection and face-matching algorithms. Such computer algorithms may ideally help current and future projects involved with preparedness, disaster response, family reunification, and medical assistance, including Sahana Vesuvius and its manifestation as NLM’s People Locator (PL).

To assist with algorithm training/testing and in establishing “ground truth”, you are asked to annotate images of faces with drawn rectangles. A web tool, “ImageStats”, has been built for this, using part of the Vesuvius/PL infrastructure and additional open-source components. A customized-for-GCI version is available at:
http://lhce-pl-web01.nlm.nih.gov/imagestats/
Over 40,000 public images are available, but the work here will be subdivided into GCI-sized tasks.

A typical task would involve approximately 600 images, each of which has one or more faces.

For each face, you must ensure that there is an appropriately-placed rectangle around the visible portions of each eye, each ear, the nose, and the mouth, with the rectangle’s color marking which of those it is. Some judgement can be involved.

Go to the Melange task list, pick an available ImageStats task, and note the “Event” name there.

IMPORTANT: In a Melange comment when claiming your first ImageStats task, include your email address. This will speed up sending you your first account username & password.

It is not necessary to include your email address in any subsequent additional ImageStats task claims, as long as you use the same Melange account.

The ImageStats team needs your email address to register you. Get it to us (but not before you have decided on a task) either by:

• Including it in a Melange comment as indicated above (recommended);
• Sending email to imagestatsaccount@nlm.nih.gov, with the subject line “Register for ImageStats Tasks”.

You will be sent back a user name and password, that the mentor will associate with your particular task. Be patient, this is a manual process. Ideally, this response happens before the mentor grants your claim.

Go to http://lhce-pl-web01.nlm.nih.gov/imagestats/ and log-on with your registered user ID and password.

• Specific Instructions - Face Landmark Annotation Task

Use Melange as before. In a comment, it will be helpful to mention that you already have supplied your email address. The mentor will then associate your new task with your previous credentials, and send you a different username and password specific to the new task, ideally prior to granting your claim.

Send email to discuss@sahanafoundation.org

The GCI tasks draw from the “Annotated Facial Landmarks in the Wild” set of over 40,000 public images. As the name implies, there are some annotations available for this set, but these are of a different style than what we need.

The original version was rapid-prototyped from a number of open source components, among them Vesuvius, GWT, GXT, vaadin, Tomcat, mysql, and SOLR. The instance is hosted and managed using NLM’s Vesuvius/PL infrastructure. It was first adapted to Google Code In in 2012, where it was successfully used for overall-face and skin-tone patch annotations.

Recognizing that this version of ImageStats was overly complicated in structures and dependencies, it was reimplementation in a simpler manner as a Google Summer of Code 2013 project (thanks, Mohit and Ajay).

Finally, both versions were tweaked to handle the new task requirements of GCI 2013. The original version is more stable and complete, and will be used at the outset of GCI 2013. The newer version may be used later.

Source code for these versions are available here.