1 |
The Impact of Removing Head Movements on Audio-visual Speech Enhancement
|
|
|
|
In: ICASSP 2022 - IEEE International Conference on Acoustics, Speech and Signal Processing ; https://hal.inria.fr/hal-03551610 ; ICASSP 2022 - IEEE International Conference on Acoustics, Speech and Signal Processing, IEEE Signal Processing Society, May 2022, Singapore, Singapore. pp.1-5 (2022)
|
|
BASE
|
|
Show details
|
|
2 |
Expression-preserving face frontalization improves visually assisted speech processing ...
|
|
|
|
BASE
|
|
Show details
|
|
3 |
Robust Face Frontalization For Visual Speech Recognition
|
|
|
|
In: International Conference on Computer Vision Workshops ; https://hal.inria.fr/hal-03326002 ; International Conference on Computer Vision Workshops, IEEE, Oct 2021, Montreal - Virtual, Canada ; http://iccv2021.thecvf.com/home (2021)
|
|
BASE
|
|
Show details
|
|
4 |
Robust Face Frontalization For Visual Speech Recognition
|
|
|
|
In: ICCV 2021 - International Conference on Computer Vision Workshops ; https://hal.inria.fr/hal-03326002 ; ICCV 2021 - International Conference on Computer Vision Workshops, IEEE, Oct 2021, Montreal - Virtual, Canada. pp.1-16 ; http://iccv2021.thecvf.com/home (2021)
|
|
Abstract:
International audience ; Face frontalization consists of synthesizing a frontallyviewed face from an arbitrarily-viewed one. The main contribution of this paper is a robust frontalization method that preserves non-rigid facial deformations, i.e. expressions, to perform lip reading. The method iteratively estimates the rigid transformation (scale, rotation, and translation) and the non-rigid deformation between 3D landmarks extracted from an arbitrarily-viewed face, and 3D vertices parameterized by a deformable shape model. An important merit of the method is its ability to deal with large Gaussian and non-Gaussian errors in the data. For that purpose, we use the generalized Student-t distribution. The associated EM algorithm assigns a weight to each observed landmark, the higher the weight the more important the landmark, thus favoring landmarks that are only affected by rigid head movements. We propose to use the zero-mean normalized cross-correlation (ZNCC) score to evaluate the ability to preserve facial expressions. We show that the method, when incorporated into a deep lipreading pipeline, considerably improves the word classification score on an in-the-wild benchmark.
|
|
Keyword:
[INFO.INFO-CV]Computer Science [cs]/Computer Vision and Pattern Recognition [cs.CV]
|
|
URL: https://hal.inria.fr/hal-03326002v3/file/Kang-ICCV21W-HAL.pdf https://hal.inria.fr/hal-03326002 https://hal.inria.fr/hal-03326002v3/document
|
|
BASE
|
|
Hide details
|
|
5 |
Robust Face Frontalization For Visual Speech Recognition
|
|
|
|
In: ICCV 2021 - International Conference on Computer Vision Workshops ; https://hal.inria.fr/hal-03326002 ; ICCV 2021 - International Conference on Computer Vision Workshops, IEEE, Oct 2021, Montreal - Virtual, Canada. pp.1-16 ; http://iccv2021.thecvf.com/home (2021)
|
|
BASE
|
|
Show details
|
|
6 |
Narrow-band Deep Filtering for Multichannel Speech Enhancement
|
|
|
|
In: https://hal.inria.fr/hal-02378413 ; 2020 (2020)
|
|
BASE
|
|
Show details
|
|
7 |
Face Frontalization Based on Robustly Fitting a Deformable Shape Model to 3D Landmarks
|
|
|
|
In: https://hal.archives-ouvertes.fr/hal-02980346 ; 2020 (2020)
|
|
BASE
|
|
Show details
|
|
8 |
Voice Activity Detection Based on Statistical Likelihood Ratio With Adaptive Thresholding
|
|
|
|
In: IWAENC 2016 - International Workshop on Acoustic Signal Enhancement (IWAENC) ; https://hal.inria.fr/hal-01349776 ; IWAENC 2016 - International Workshop on Acoustic Signal Enhancement (IWAENC), Sep 2016, Xi'an, China. pp.1-5, ⟨10.1109/IWAENC.2016.7602911⟩ (2016)
|
|
BASE
|
|
Show details
|
|
|
|