We are looking for a developer with experience in face detection/recognition to build a face comparison system.
The system should be able to:
~ Take as input TWO photos.
~ Confirm that each photo contains a SINGLE face.
~ COMPARE the two faces to judge whether or not it is the same person.
Applicants should give brief details of one or more face detection/recognition projects they have completed.
Est. Time: 1 to 3 months, 30+ hrs/week
We are building a UI composed of 2 side-by-side landscape 16:9 touchscreens that are displaying a video of various items advancing at a given speed (x m/s) on a conveyor. The video signal comes from a linescan (Sick from Sweden) camera that sends only lines (one pixel wide/width of conveyor high) packets to the PC. These lines need to be collaged together and displayed in real-time on the two screens and give the impression that we are watching a video of items on a conveyor passing in front of us. The operator uses his finger(s) or stylet to "pick" items. At picking time an "X" appears over the picked item. Picking with one finger will mark the item with a certain color and with 2 fingers with another color. The "X" needs to track at the same speed as the conveyor so it stays over the picked item in the video (an encoder signal provide speed). An analysis is performed around the X to find the item contour. The contour is drawn and tracks as well on the screen. A bitmap (sized by the contour) of the item is saved to a database.
We will start developing with normal monitors (no touchscreens) just a mouse (right and left clicks). The linescan video packets will not come from a live camera but from a harddrive. We will supply this data. The best video card for the job will need to be determined jointly.
This project's objective is to use iOS devices cameras to read MRZ codes from passports (https://en.wikipedia.org/wiki/Machine-readable_passport).
We created a project using opencv and tesseract, got good results, but not good enough.