Olfactory Summer Studentship

So all the exams are over, it's all finished (this year anyway!), and I've been looking for a job.

I have been hoping to work in the department after doing so last year, and heard of a scholarship that was up for grabs to do with olfactory tagging of photos.

Anyway, I applied a while ago and managed to get an interview yesterday (after my last exam and before the pubbage). I heard back earlier today and I have been offered the job, it's a ten week scholarship in the department which pays £240 a week. The main idea is to tag photos with smells, much in the same way that flickr.com uses words to tag photos. I'm not set to start until the 1st of July but thought I would post some early details I have on the project.

Below is an extract from the outline of the project, defining exactly what it is about:

In order to investigate the use of olfaction as part of HCI, the work undertaken during the studentship will centre on the construction of an application which will be intended to both exploit the advantages of olfaction (memory and learning), and apply these advantages to a current problem in human computer interaction — the browsing of large non-textual media collections such as digital photographs. The design and construction of this application will contribute to the general aims of the studentship to improve our understanding of olfaction in HCI and the expected outcomes which are described in detail in the following sections.

The problems of browsing large collections of non-textual digital media are well known, with ineffective tools currently available for browsing and searching such collections. Most of the proposed methods to overcome the problems of browsing and searching involve the tagging of individual, or groups of photographs, with keywords which can be later used for retrieval. However, other research has found that many people find such tagging of photographs inconvenient and many do not store digital photographs in any systematic way. In any event, such a system may still break down if inconsistent labels are given to different photographs.

The application to be constructed as part of this studentship will build on existing research with Tangible User Interfaces (TUIs), to allow users to build search queries over digital photograph collections via RFID (Radio Frequency Identification) tags attached to 'smell cubes' (small plastic cubes containing a fragrance impregnated in a cotton wool pad — www.daleair.com). This would allow users to associate smells easily and quickly with collections of photographs by moving the appropriate cube close to the RFID tag reader, and would help to make the process of recall easier by using the powerful memory effects of smell to remind users of the environment in which they took the photographs and thus aid their recall. For example, photographs of a holiday in India may be quickly associated with the smell of spices. In years to come the user may be looking for a particular photograph of a friend. The user can remember the friend was standing outside a building but cannot quite remember where or when the photograph was taken (such attributes being the current way in which images are tagged). To find the photograph the user would waft each smell cube under their nose. On using the spice smell cube, the user is transported back to their holiday in India and can remember where the photograph was taken. On passing the smell cube (with attached RFID tag) over the RFID sensor, the system filters the photographs on screen to show only a small subset, one of which is the photograph desired by the user — the users' friend standing outside the building.

The work will be broken down in the 10 weeks as follows:

Anyway, I'm quite pleased with getting the scholarship and look forward to an interesting summer working on it. Sorry to the other guys who were in for it (Dave et al), I'm sure you'll all find something soon enough.
- Chris