Event Location: NSH 1507
Bio: Dr. Meirav Galun received the BSc degree in mathematics and computer science
from Tel Aviv University in 1985 and her MSc and Phd degrees in applied
mathematics from the Weizmann Institute of Science in 1992 and 1998,
respectively. She is currently a research scientist in the computer science
and applied mathematics department at the Weizmann Institute of Science.
Dr. Galun’s research has focused on computer vision, data analysis and
medical imaging. She has developed multiscale algorithms for image
segmentation, edge detection and clustering.

Abstract: Images and data are multiscale by nature: significant data structures
(edges, regions, shapes, clusters) are often found at different scales.
Local or uni-scale features are usually insufficient to identify and extract
such structures, whereas appropriate multiscale features can explain the
nature of the data. Full perception is often obtained only when integrating
information across multiple scales. In this talk, I will present multiscale
methods that we have developed for analysis and perception of complex data.
I will show the power of our multiscale approach on a variety of data and
image analysis problems, including edge detection in very noisy images,
segmentation of complex imagery, data clustering, and shape classification.