{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,6]],"date-time":"2026-05-06T10:50:02Z","timestamp":1778064602297,"version":"3.51.4"},"reference-count":26,"publisher":"Springer Science and Business Media LLC","issue":"1","license":[{"start":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T00:00:00Z","timestamp":1645660800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T00:00:00Z","timestamp":1645660800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100005711","name":"Universit\u00e4t Hamburg","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100005711","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["K\u00fcnstl Intell"],"published-print":{"date-parts":[[2022,3]]},"abstract":"<jats:title>Abstract<\/jats:title><jats:p>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\u2019s 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<\/jats:p>","DOI":"10.1007\/s13218-021-00746-2","type":"journal-article","created":{"date-parts":[[2022,2,24]],"date-time":"2022-02-24T10:03:02Z","timestamp":1645696982000},"page":"91-98","update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":10,"title":["Multi-phase Fine-Tuning: A New Fine-Tuning Approach for Sign Language Recognition"],"prefix":"10.1007","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1545-9346","authenticated-orcid":false,"given":"Noha","family":"Sarhan","sequence":"first","affiliation":[]},{"given":"Mikko","family":"Lauri","sequence":"additional","affiliation":[]},{"given":"Simone","family":"Frintrop","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2022,2,24]]},"reference":[{"key":"746_CR1","doi-asserted-by":"crossref","unstructured":"Azizpour H et al (2015) From generic to specific deep representations for visual recognition. In: CVPR DeepVision workshop, pp 36\u201345","DOI":"10.1109\/CVPRW.2015.7301270"},{"key":"746_CR2","unstructured":"Bengio Y (2012) Deep learning of representations for unsupervised and transfer learning. In: ICML workshop on unsupervised and transfer learning, pp 17\u201336"},{"key":"746_CR3","doi-asserted-by":"crossref","unstructured":"Camgoz NC et al (2020) Sign language transformers: joint end-to-end sign language recognition and translation. In: CVPR, pp 10023\u201310033","DOI":"10.1109\/CVPR42600.2020.01004"},{"key":"746_CR4","doi-asserted-by":"crossref","unstructured":"Cao C, Zhang Y, Wu Y, Lu H, Cheng J (2017) Egocentric gesture recognition using recurrent 3d convolutional neural networks with spatiotemporal transformer modules. In: ICCV, pp 3763\u20133771","DOI":"10.1109\/ICCV.2017.406"},{"key":"746_CR5","doi-asserted-by":"crossref","unstructured":"Cihan Camgoz N et al (2017) Subunets: end-to-end hand shape and continuous sign language recognition. In: ICCV, pp 3056\u20133065","DOI":"10.1109\/ICCV.2017.332"},{"key":"746_CR6","doi-asserted-by":"crossref","unstructured":"Cui R, Liu H, Zhang C (2017) Recurrent convolutional neural networks for continuous sign language recognition by staged optimization. In: CVPR, pp 1\u20134","DOI":"10.1109\/CVPR.2017.175"},{"key":"746_CR7","doi-asserted-by":"crossref","unstructured":"Daroya R, Peralta D, Naval P (2018) Alphabet sign language image classification using deep learning. In: TENCON 2018-2018 IEEE region 10 conference, IEEE, pp 0646\u20130650","DOI":"10.1109\/TENCON.2018.8650241"},{"key":"746_CR8","unstructured":"Forster J et al (2014) Extensions of the sign language recognition and translation corpus RWTH-PHOENIX-weather. In: International conference on language resources and evaluation, pp 1911\u20131916"},{"key":"746_CR9","unstructured":"Glorot X, Bengio Y (2010) Understanding the difficulty of training deep feedforward neural networks. In: International conference on artificial intelligence and statistics, pp 249\u2013256"},{"key":"746_CR10","doi-asserted-by":"publisher","first-page":"108","DOI":"10.1016\/j.cviu.2015.09.013","volume":"141","author":"O Koller","year":"2015","unstructured":"Koller O, Forster J, Ney H (2015) Continuous sign language recognition: towards large vocabulary statistical recognition systems handling multiple signers. Comput Vis Image Underst 141:108\u2013125","journal-title":"Comput Vis Image Underst"},{"key":"746_CR11","doi-asserted-by":"crossref","unstructured":"Koller O et al (2016) Deep sign: hybrid CNN-HMM for continuous sign language recognition. In: BMVC, pp 136.1\u2013136.12","DOI":"10.5244\/C.30.136"},{"key":"746_CR12","doi-asserted-by":"crossref","unstructured":"Koller O et al (2017) Re-sign: re-aligned end-to-end sequence modelling with deep recurrent CNN-HMMS. In: CVPR, pp 4297\u20134305","DOI":"10.1109\/CVPR.2017.364"},{"key":"746_CR13","doi-asserted-by":"crossref","unstructured":"Li D et al (2020) Word-level deep sign language recognition from video: a new large-scale dataset and methods comparison. In: IEEE WACV, pp 1459\u20131469","DOI":"10.1109\/WACV45572.2020.9093512"},{"key":"746_CR14","unstructured":"Montone G, O\u2019Regan JK, Terekhov AV (2015) The usefulness of past knowledge when learning a new task in deep neural networks. In: CoCo@ NIPS, pp 10\u201318"},{"key":"746_CR15","unstructured":"Montone G, O\u2019Regan JK, Terekhov AV (2017) Gradual tuning: a better way of fine tuning the parameters of a deep neural network. arXiv preprint arXiv:1711.10177"},{"key":"746_CR16","doi-asserted-by":"crossref","unstructured":"Penatti OA, Nogueira K, dos Santos JA (2015) Do deep features generalize from everyday objects to remote sensing and aerial scenes domains? In: CVPR workshops, pp 44\u201351","DOI":"10.1109\/CVPRW.2015.7301382"},{"key":"746_CR17","doi-asserted-by":"crossref","unstructured":"Pigou L et al (2014) Sign language recognition using convolutional neural networks. In: ECCV workshops ChaLearn looking at people, pp 572\u2013578","DOI":"10.1007\/978-3-319-16178-5_40"},{"issue":"3","key":"746_CR18","doi-asserted-by":"publisher","first-page":"211","DOI":"10.1007\/s11263-015-0816-y","volume":"115","author":"O Russakovsky","year":"2015","unstructured":"Russakovsky O et al (2015) Imagenet large scale visual recognition challenge. Int J Comput Vis 115(3):211\u2013252","journal-title":"Int J Comput Vis"},{"key":"746_CR19","doi-asserted-by":"crossref","unstructured":"Sarhan N, Frintrop S (2020) Transfer learning for videos: from action recognition to sign language recognition. In: IEEE ICIP, IEEE, pp 1811\u20131815","DOI":"10.1109\/ICIP40778.2020.9191289"},{"key":"746_CR20","doi-asserted-by":"crossref","unstructured":"Sharif Razavian A et al (2014) CNN features off-the-shelf: an astounding baseline for recognition. In: CVPR workshop, pp 806\u2013813","DOI":"10.1109\/CVPRW.2014.131"},{"key":"746_CR21","doi-asserted-by":"crossref","unstructured":"Szegedy C et al (2015) Going deeper with convolutions. In: CVPR, pp 1\u20139","DOI":"10.1109\/CVPR.2015.7298594"},{"key":"746_CR22","doi-asserted-by":"crossref","unstructured":"Szegedy C et al (2016) Rethinking the inception architecture for computer vision. In: CVPR, pp 2818\u20132826","DOI":"10.1109\/CVPR.2016.308"},{"issue":"5","key":"746_CR23","doi-asserted-by":"publisher","first-page":"1299","DOI":"10.1109\/TMI.2016.2535302","volume":"35","author":"N Tajbakhsh","year":"2016","unstructured":"Tajbakhsh N et al (2016) Convolutional neural networks for medical image analysis: full training or fine tuning? IEEE Trans Med Imaging 35(5):1299\u20131312","journal-title":"IEEE Trans Med Imaging"},{"issue":"6","key":"746_CR24","doi-asserted-by":"publisher","first-page":"821","DOI":"10.18178\/ijmlc.2019.9.6.879","volume":"9","author":"LKS Tolentino","year":"2019","unstructured":"Tolentino LKS et al (2019) Static sign language recognition using deep learning. Int J Mach Learn Comput 9(6):821\u2013827","journal-title":"Int J Mach Learn Comput"},{"key":"746_CR25","doi-asserted-by":"publisher","first-page":"7957","DOI":"10.1007\/s00521-019-04691-y","volume":"32","author":"A Wadhawan","year":"2020","unstructured":"Wadhawan A, Kumar P (2020) Deep learning-based sign language recognition system for static signs. Neural Comput Appl 32:7957\u20137968","journal-title":"Neural Comput Appl"},{"key":"746_CR26","unstructured":"Yosinski J, Clune J, Bengio Y, Lipson H (2014) How transferable are features in deep neural networks? In: NIPS, pp 3320\u20133328"}],"container-title":["KI - K\u00fcnstliche Intelligenz"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13218-021-00746-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s13218-021-00746-2\/fulltext.html","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s13218-021-00746-2.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2022,5,6]],"date-time":"2022-05-06T08:08:21Z","timestamp":1651824501000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s13218-021-00746-2"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2022,2,24]]},"references-count":26,"journal-issue":{"issue":"1","published-print":{"date-parts":[[2022,3]]}},"alternative-id":["746"],"URL":"https:\/\/doi.org\/10.1007\/s13218-021-00746-2","relation":{},"ISSN":["0933-1875","1610-1987"],"issn-type":[{"value":"0933-1875","type":"print"},{"value":"1610-1987","type":"electronic"}],"subject":[],"published":{"date-parts":[[2022,2,24]]},"assertion":[{"value":"24 April 2020","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"22 November 2021","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"24 February 2022","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}}]}}