MIT EECS - Quanta Computer Undergraduate Research and Innovation Scholar
Frederic P. Durand
Currently, handwritten videos have no tools that allow one to easily create his/her work in a non-sequential order. The user must have a plan prior to the recording and must restart a recording from scratch if s/he chooses to deviate from the plan, regardless of how small the refinement is. The teams application currently supports features that allow the video author to smoothly make insertions of text and audio without affecting the time flow. I will be working on the handwriting beautification that will allow the video authors to create videos with a more readable font. This will allow the users to focus on the content of the video rather than deal with rewriting an equation that is not readable.
During my past internship at Microsoft, I created an Outlook application which required that I use innovative tools, such as Named Entity Recognizer (NER), developed by research teams. This piqued my interest in participating in the creation of these pioneering tools, rather than exclusively applying them. Ive also had a brief introduction to the area Ill be researching in the Machine Learning class I took last semester.