Multi-phase Fine-Tuning: A New Fine-Tuning Approach for Sign Language Recognition

dc.contributor.author	Sarhan, Noha
dc.contributor.author	Lauri, Mikko
dc.contributor.author	Frintrop, Simone
dc.date.accessioned	2022-05-30T08:56:52Z
dc.date.available	2022-05-30T08:56:52Z
dc.date.issued	2022
dc.description.abstract	In this paper, we propose multi-phase fine-tuning for tuning deep networks from typical object recognition to sign language recognition (SLR). It extends the successful idea of transfer learning by fine-tuning the network’s weights over several phases. Starting from the top of the network, layers are trained in phases by successively unfreezing layers for training. We apply this novel training approach to SLR, since in this application, training data is scarce and differs considerably from the datasets which are usually used for pre-training. Our experiments show that multi-phase fine-tuning can reach significantly better accuracy in fewer training epochs compared to previous fine-tuning techniques	de
dc.identifier.doi	10.1007/s13218-021-00746-2
dc.identifier.pissn	1610-1987
dc.identifier.uri	http://dx.doi.org/10.1007/s13218-021-00746-2
dc.identifier.uri	https://dl.gi.de/handle/20.500.12116/38669
dc.publisher	Springer
dc.relation.ispartof	KI - Künstliche Intelligenz: Vol. 36, No. 1
dc.relation.ispartofseries	KI - Künstliche Intelligenz
dc.subject	Sign language recognition
dc.subject	Transfer learning
dc.title	Multi-phase Fine-Tuning: A New Fine-Tuning Approach for Sign Language Recognition	de
dc.type	Text/Journal Article
gi.citation.endPage	98
gi.citation.startPage	91

Sammlungen