Last week, I met with Ann Zawistoski to refine our search for articles on facial recognition software. From that meeting, I was able to find a couple sources that look very promising for giving us the techniques and tools we need for identifying and re-identifying faces. Then on Thursday, we met again with Nat Wilson to go over our findings and discuss the best tools and sources we’ve found so far. This week, we’ll be working on looking through and ranking those tools we discussed on Thursday, as well as any others we find online that look promising. We’ll be using a set of criteria established over the course of the past week (things like ease of use, ability to do batch imports/exports of data, ability to maintain privacy, etc.). That way, we should have a concrete way to quanititatively determine the best tools by next week, with any luck. We’re looking to narrow down the tools we have to about 5-7 finalists, and then we can further narrow it down from there.
Currently, Sahree and I are working with Nat Wilson in the archives department to review existing facial recognition software tools. The end goal of this project is to be able to use a tool to recognize and label pictures of alumni with much greater speed and accuracy than we currently have by doing the process manually.To that end, the three of us are reviewing some articles this week dealing with exisiting tools. In particular, Nat gave us a list of 50 or so facial recognition tools (the list was current as of late 2013), and I highlighted five or so that I thought would be particularly useful. I tried to focus on tools that were free, independent of outside software (so we don’t have to send private information to external networks), and simple to use products designed to work on their own (as opposed to being part of a code library for use in developing larger applications).