In this paper we present the application of hidden conditional random fields (HCRFs) to modeling speech for visual speech recognition. HCRFs may be easily adapted to model long range dependencies across an observation sequence. As a result visual word recognition performance can be improved as the model is able to take more of a contextual approach to generating state sequences. Results are presented from a speaker-dependent, isolated digit, visual speech recognition task using comparisons with a baseline HMM system. We firstly illustrate that word recognition rates on clean video using HCRFs can be improved by increasing the number of past and future observations being taken into account by each state. Secondly we compare model performances using various levels of video compression on the test set. As far as we are aware this is the first attempted use of HCRFs for visual speech recognition.
|Title of host publication
|Proceedings 13th International Machine Vision and Image Processing Conference, 2009
|Subtitle of host publication
|Ken Dawson-Howe, Rozenn Dahyot, Anil Kokaram, Gerard Lacey
|Place of Publication
|Los Alamitos, Calif.
|Number of pages
|Published - 2009
|13th International Machine Vision and Image Processing Conference - Dublin, Ireland
Duration: 2 Sept 2009 → 4 Sept 2009
|13th International Machine Vision and Image Processing Conference
|2/09/09 → 4/09/09