“Electrical engineering” often spurs images of semi-conductors and computer chips, but not for Prof. Sheila Hemami, electrical and computer engineering, who concentrates on the systems aspect of electrical engineering. Her research focuses on signal processing, in which she looks at how information can be most reliably transmitted.

Hemami and her colleagues have dealt with how the American Sign Language Video for the American deaf community can be improved so that the deaf community could also take advantage of the cellular revolution.

The most important aspect of this communications device – similar to video chat on a cell phone – is the quality of the signer images versus that of the backdrop images.

The next most important part of the video is the signer’s face. A signer’s face plays an integral role in American Sign Language, as minute differences, such as raised eyebrows versus eyebrows at their normal position, can mean the difference between a question and a statement.

According to Hemami, these small movements, which are rarely important in a conversation between people with perfect hearing, play big roles in conversations between deaf people. Currently, the signs are hard to decipher with the communications technology.

“We know that you can have a high definition video, and depending on the bandwidth, you can always change the resolution,” explained Hemami. 

The problem is that the network connection bandwidth on the current data network is ten times lower than the quality needed for Sign Language Video to be intelligible if it were coded using current approaches. With a very distorted video frame, the sender’s meaning is lost to the receiver.

Now, Hemami hopes to take her research another step forward by contributing to a large, multi-faculty project, aiming to make “video delivery more efficient through mobile devices.” The project - which is currently being finalized - will involve four faculty in the School of Electrical and Computer Engineering.

 “The industry predicts that in the next five years, the amount video being transmitted over the network – so to your computer, to your ipad, your phone, anything – will go up by a factor of sixty-five.” 

However, the current Internet was not designed to be flexible enough to handle the heavy traffic of data that accompanies increased use of video. 

This project aims to make video more available to all users. Hemami will contribute to this project by incorporating human perception of video into a network's routing algorithms.

“Every engineer wants to build something,” said Hemami expressing her and her colleague’s hopes. Hemami’s own aspiration to become an electrical and computer engineer was inspired by her father – also an electrical engineer. Though her father works with biomedical control, he exposed her to engineering early in her life.

She majored in electrical engineering at the University of Michigan, and received her doctorate in the field at Stanford University.  

An engineering degree, Hemami explained, is not simply a certificate, but a guide that helps to solve a wide variety of open-ended questions. According to Hemami, an engineer will solve complex problems in a “methodological and efficient way.” According to Hemami, though others may give up and deem believe that hard questions are impossible to solve, the engineer will break up the question, simplify the problem, and work at it until he or she finds the answer.

“Everybody should be an engineer.” Hemami encouraged. “The world would be a better place if everyone was a better problem solver.”