Middlebury College
Department of Computer Science Seminar
Friday, December 9 at 12:15 PM:
Neural Networks for Handwriting Recognition
Niraj Dhungel '06
The fact that the human brain performs some operations with repeated accuracy prompted the Artificial Neural Network model. Artificial Neural Networks or simply Neural Networks attempt to simulate the operations of the human brain and have proved particularly successful in pattern recognition.
Handwriting recognition is a pattern recognition problem that neural networks have been very successful in solving. Neural networks have demonstrated high accuracy in recognizing handwritten text, be it handwriting scanned previously or while it is being written. In this talk I will present two different neural networks—Convolutional neural network and Time Delay neural network—that have had particular success in offline and online handwriting recognition, respectively.
Friday, December 9 at 12:45 PM:
MPEG And Beyond: A Look At Video Compression Methods
Duncan Sanford '06
How much do we really see, and how much do our eyes ignore? What's the difference between a VHS cassette and a DVD? The science of video compression looks at how much space—or how little—we need to store and transmit recognizable video data. In this talk I will explore color space, motion estimation, the discrete cosine transform, and other techniques used in several popular image and video compression standards.
Friday, December 9, 2005
12:15 p.m. to 1:15 p.m.
McCardell Bicentennial Hall 538
Lunch will be provided at 12:05 p.m.
All are welcome to attend!
This event is supported by the Computer Science Enrichment Fund