Département Image, Données, Signal collection |
 |
Conference papers
Learning Visual Voice Activity Detection with an Automatically Annotated Dataset
Abstract : Visual voice activity detection (V-VAD) uses visual features to predict whether a person is speaking or not. V-VAD is useful whenever audio VAD (A-VAD) is inefficient either because the acoustic signal is difficult to analyze or is simply missing. We propose two deep architectures for V-VAD, one based on facial landmarks and one based on optical flow. Moreover, available datasets, used for learning and for testing V-VAD, lack content variability. We introduce a novel methodology to automatically create and annotate very large datasets in-the-wild, based on combining A-VAD and face detection. A thorough empirical evaluation shows the advantage of training the proposed deep V-VAD models with such a dataset.
Cited literature [31 references]
https://hal.inria.fr/hal-02882229
Contributor : Team Perception <>
Submitted on : Wednesday, September 23, 2020 - 11:39:19 AM
Last modification on : Tuesday, November 24, 2020 - 4:00:09 PM
Contributor : Team Perception <>
Submitted on : Wednesday, September 23, 2020 - 11:39:19 AM
Last modification on : Tuesday, November 24, 2020 - 4:00:09 PM
File
GUY_ICPR2020_sub.pdf
Files produced by the author(s)