2008 8th IEEE International Conference on Automatic Face &Amp; Gesture Recognition 2008
DOI: 10.1109/afgr.2008.4813439
|View full text |Cite
|
Sign up to set email alerts
|

Efficient approximations to model-based joint tracking and recognition of continuous sign language

Abstract: We propose several tracking adaptation approaches to recover from early tracking errors in sign language recognition by optimizing the obtained tracking paths w.r.t. to the hypothesized word sequences of an automatic sign language recognition system. Hand or head tracking is usually only optimized according to a tracking criterion. As a consequence, methods which depend on accurate detection and tracking of body parts lead to recognition errors in gesture and sign language processing. We analyze an integrated … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
3
0
2

Year Published

2009
2009
2017
2017

Publication Types

Select...
6
3

Relationship

2
7

Authors

Journals

citations
Cited by 23 publications
(5 citation statements)
references
References 14 publications
0
3
0
2
Order By: Relevance
“…In many state-of-the-art works on sign language recognition and analysis, the locations of the articulating hands are described with a single (x, y)-coordinate pair each, typically aimed at indicating the positions of the palms [2]. This level of presentation has also been used for annotating publicly available sign language video databases and benchmarks [3].…”
Section: Head Occlusions In Signingmentioning
confidence: 99%
“…In many state-of-the-art works on sign language recognition and analysis, the locations of the articulating hands are described with a single (x, y)-coordinate pair each, typically aimed at indicating the positions of the palms [2]. This level of presentation has also been used for annotating publicly available sign language video databases and benchmarks [3].…”
Section: Head Occlusions In Signingmentioning
confidence: 99%
“…the tracking, feature extraction and statistical visual-phonetic modeling. The lack of phonetic transcriptions, standardized phonetic models and lexica for SL corpora render continuous Sign Language Recognition (SLR) quite difficult [19,16]. SLR tasks are found even more demanding due to the variability of continuous signing characteristics and the multiple information streams, as for instance handshape and movement.…”
Section: Introductionmentioning
confidence: 99%
“…Multiple cues are combined in [10] for isolated sign recognition. Another aspect concerns continuous SLR [16,18,19,20] and issues such as coarticulation and movement epenthesis. Nested dynamic programming is employed in [18] to handle movement epenthesis.…”
Section: Introductionmentioning
confidence: 99%
“…Los video publicados sólo tiene 195*165 pixeles según se comenta en la documentación. En [44] se presenta una variante de la base de datos titulada "RWTH-Boston-104"donde el objetivo estuvo puesto en el reconocimiento continuo. Contiene 161 sentencias con 104 señas distintas.…”
Section: Bases De Datos De Lengua De Señasunclassified