Monday, September 19, 2011

Compressed Sensing and other stuff

Hey everyone..

I have registered for FiO 2011, and have started looking into the schedule. There are many events this year and I don’t want to put off looking at the schedule till the last moment and miss out on the more interesting ones. 

The first day, Sunday 10/16, 4-6pm, seems to have a very good What’s Hot in Optics session that should be interesting. Usually a start-off session like this provides a very good overview of things to come, so I will surely be there. 

Sunday evening (7.30-8.30pm) has meetings for OSA technical groups. I had signed up for a couple for groups and got emails from them inviting anyone interested to join the meetings. There should be discussions for areas like Optical Design and Instrumentation, Holography, Information Acquisition, Biomedical Optics, etc. Meetings like these help people get involved with OSA in these technical areas.

Monday has a great session on Compressed Sensing (FMM, from 4-6pm). Rebecca Willet from Duke is going to have a 45 minute tutorial style talk on compressed sensing for optical imaging; she mentions IR and focal plane arrays in her abstract. There is much work being done to minimize the data being sensed as well as transmitted. Her group works on compressed sensing, exploiting sparsity in the nature of data to obtain compression. She has worked on compression in the presence of Poisson noise, coded aperture imaging, multi-aperture imaging for thin imagers, and more. So the tutorial talk should be a good overview.

There are also talks from George Barbastathis' group at MIT, and some from David Brady’s group in this session that should be interesting. There is one by K. MacCabe that takes advantage of aliasing high spatial frequencies, folding them into the lower spatial frequency region which is more resistant to loss during defocus. In my opinion, turning around aliasing to your advantage is always brilliant! This session certainly promises to be very interesting.

Oh.. I just found an old video of Rebecca Willet giving a talk at some CS workshop at Duke; summarizes compressed sensing to some extent. Here are the slides for her talk, do check out the slides because the video below does not display them. Also, audio-video here are a bit out of sync, but with the slides it's a pretty good talk. Her talk at FiO will probably have more recent stuff, but this should give you a flavour of what to expect at the CS session.



Ooh, just spotted this, Kevin Kelly from Rice seems to have a 30 minute talk (LTuC3) Tuesday morning, on their work on single pixel compressive imaging where they use a fast DMD to time multiplex and compress the object information and then capture only the minimum required on a single pixel! 

That's it for now. I'll post more about other sessions and events over the next few weeks. Cheers and thanks for reading!


  1. Hi! Thanks for pointing out that the Tech Division meetings were not showing on the online program planner. They have now been added and are available to add to your itinerary.