Abstract: Visual Speech Recognition (VSR), also known as lip-reading, requires accurate mouth tracking and alignment to achieve high performance. Conventional approaches rely on complex pipelines ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results