Zobrazeno 1 - 10
of 184
pro vyhledávání: '"Gloss-free"'
Sign language translation (SLT) is a challenging task that involves translating sign language images into spoken language. For SLT models to perform this task successfully, they must bridge the modality gap and identify subtle variations in sign lang
Externí odkaz:
http://arxiv.org/abs/2411.16789
Autor:
Chen, Zhigang, Zhou, Benjia, Huang, Yiqing, Wan, Jun, Hu, Yibo, Shi, Hailin, Liang, Yanyan, Lei, Zhen, Zhang, Du
Sign Language Representation Learning (SLRL) is crucial for a range of sign language-related downstream tasks such as Sign Language Translation (SLT) and Sign Language Retrieval (SLRet). Recently, many gloss-based and gloss-free SLRL methods have bee
Externí odkaz:
http://arxiv.org/abs/2408.09949
Gloss-free sign language translation (SLT) aims to develop well-performing SLT systems with no requirement for the costly gloss annotations, but currently still lags behind gloss-based approaches significantly. In this paper, we identify a representa
Externí odkaz:
http://arxiv.org/abs/2405.14312
Automatic Sign Language Translation requires the integration of both computer vision and natural language processing to effectively bridge the communication gap between sign and spoken languages. However, the deficiency in large-scale training data t
Externí odkaz:
http://arxiv.org/abs/2405.04164
Autor:
Chen, Zhigang, Zhou, Benjia, Li, Jun, Wan, Jun, Lei, Zhen, Jiang, Ning, Lu, Quan, Zhao, Guoqing
Previous Sign Language Translation (SLT) methods achieve superior performance by relying on gloss annotations. However, labeling high-quality glosses is a labor-intensive task, which limits the further development of SLT. Although some approaches wor
Externí odkaz:
http://arxiv.org/abs/2403.12556
Autor:
Pedro Dal Bianco, Gastón Ríos, Waldo Hasperué, Oscar Stanchi, Facundo Quiroga, Franco Ronchetti
Publikováno v:
Journal of Computer Science and Technology, Vol 24, Iss 2 (2024)
Sign Language Translation (SLT) is a challenging task due to its cross-domain nature, different grammars and lack of data. Currently, many SLT models rely on intermediate gloss annotations as outputs or latent priors. Glosses can help models to corre
Externí odkaz:
https://doaj.org/article/2737799e69fc4902bfbfc019b1a62327
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Gloss-free Sign Language Production (SLP) offers a direct translation of spoken language sentences into sign language, bypassing the need for gloss intermediaries. This paper presents the Sign language Vector Quantization Network, a novel approach to
Externí odkaz:
http://arxiv.org/abs/2309.12179
Autor:
Zhou, Benjia, Chen, Zhigang, Clapés, Albert, Wan, Jun, Liang, Yanyan, Escalera, Sergio, Lei, Zhen, Zhang, Du
Sign Language Translation (SLT) is a challenging task due to its cross-domain nature, involving the translation of visual-gestural language to text. Many previous methods employ an intermediate representation, i.e., gloss sequences, to facilitate SLT
Externí odkaz:
http://arxiv.org/abs/2307.14768
In this paper, we tackle the problem of sign language translation (SLT) without gloss annotations. Although intermediate representation like gloss has been proven effective, gloss annotations are hard to acquire, especially in large quantities. This
Externí odkaz:
http://arxiv.org/abs/2305.12876