Making The Sky Searchable: Large Scale Astronomical Pattern Recognition
Speaker: Sam Roweis , Department of Computer Science, University of TorontoContact:
Date: December 4 2008
Time: 4:00PM to 5:00PM
Location: G449 (Patil/Kiva)
Host: Nick Roy
Nick Roy, x3-2517, email@example.comRelevant URL:
Imagine you have an uncalibrated picture of the night sky and you want to know where the telescope was pointing when the picture was taken. Since several digital catalogues are available, containing (among other data) positions and magnitudes of billions of stars, you should, in principle, be able to find source locations by analyzing the pixels of your image and then exhaustively search the catalogues and find where that pattern of sources occurs. The only catch is that the sky is pretty big and that both images and catalogues are pretty noisy. Nonetheless, by using efficient geometric hashing techniques, our group has built a universal astrometric calibration robot which, roughly speaking, takes as input a picture of the night sky and returns as output the location on the sky at which the picture was taken. This is the first step in a more ambitious effort to learn a probabilistic model which accounts for *every* image of the night sky ever taken (including all professional, amateur and historical pixels) by modeling not only astrometry but also bandpass, time, and instrument properties. I will also briefly discuss other work which applies this general research approach -- performing learning and inference in generative models trained on multiple observations -- to other domains such as proteomics, speech and signal processing.
Joint work with between University of Toronto and NYU.
Project Website: http://astrometry.net
See other events that are part of Artificial Intelligence Seminar Series 2008/2009
See other events happening in December 2008