Filters








25,953 Hits in 5.4 sec

Transferring Cross-domain Knowledge for Video Sign Language Recognition [article]

Dongxu Li, Xin Yu, Chenchen Xu, Lars Petersson, Hongdong Li
2020 arXiv   pre-print
Word-level sign language recognition (WSLR) is a fundamental task in sign language interpretation. It requires models to recognize isolated sign words from videos.  ...  Since these videos have no word-level annotation and exhibit a large domain gap from isolated signs, they cannot be directly used for training WSLR models.  ...  We thank all anonymous reviewers and ACs for their constructive comments.  ... 
arXiv:2003.03703v2 fatcat:5o7q6fd6pbcufbmzad5q4in2iy

Transferring Cross-Domain Knowledge for Video Sign Language Recognition

Dongxu Li, Xin Yu, Chenchen Xu, Lars Petersson, Hongdong Li
2020 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)  
Word-level sign language recognition (WSLR) is a fundamental task in sign language interpretation. It requires models to recognize isolated sign words from videos.  ...  Since these videos have no word-level annotation and exhibit a large domain gap from isolated signs, they cannot be directly used for training WSLR models.  ...  We thank all anonymous reviewers and ACs for their constructive comments.  ... 
doi:10.1109/cvpr42600.2020.00624 dblp:conf/cvpr/LiYXPL20 fatcat:sndlr6kwwbayzkn6yzmfj2ypsu

A Simple Multi-Modality Transfer Learning Baseline for Sign Language Translation [article]

Yutong Chen, Fangyun Wei, Xiao Sun, Zhirong Wu, Stephen Lin
2022 arXiv   pre-print
This paper proposes a simple transfer learning baseline for sign language translation. Existing sign language datasets (e.g.  ...  PHOENIX-2014T, CSL-Daily) contain only about 10K-20K pairs of sign videos, gloss annotations and texts, which are an order of magnitude smaller than typical parallel data for training spoken language translation  ...  Our work fully exploits gloss annotations for SLT by transferring within-domain knowledge from ISLR to CSLR and SLT. Sign Language Translation.  ... 
arXiv:2203.04287v1 fatcat:glaep5767jdcngfwuuymoy6wlu

Deep transfer learning base on sequenced edge grid image technique for sign language recognition

Supathep Satiman, Phayung Meesad
2022 International Journal of Health Sciences  
Researchers propose an innovative technique for video processing called Sequenced Edge Grid Images (SEGI) for sign language recognition to interpret hand gesture, body movement, and facial expression.  ...  From the results data-preprocesses technique of dataset generation and deep transfer learning was an effective way to improve the accuracy of sign language recognition.  ...  sign language recognition, and potential problems in larger video sequence data training (Y.  ... 
doi:10.53730/ijhs.v6ns5.12112 fatcat:36ubptkhaja23ivjen4dxx26qy

Transfer Learning for British Sign Language Modelling [article]

Boris Mocialov, Graham Turner, Helen Hastie
2020 arXiv   pre-print
In this paper, we examine two transfer learning techniques of fine-tuning and layer substitution for language modelling of British Sign Language.  ...  This has led to work on transfer learning methods, whereby a model developed for one language is reused as the starting point for a model on a second language, which is less resourced.  ...  Transfer Learning While transfer learning is a more general machine learning term, cross-domain adaptation of language models is used in the language modelling literature (Deena et al., 2016; Ma et al  ... 
arXiv:2006.02144v1 fatcat:7iclnhowjbfddcdvbekwozfafu

Hierarchical I3D for Sign Spotting [article]

Ryan Wong, Necati Cihan Camgöz, Richard Bowden
2022 arXiv   pre-print
Most of the vision-based sign language research to date has focused on Isolated Sign Language Recognition (ISLR), where the objective is to predict a single sign class given a short video clip.  ...  In this paper, we focus on the challenging task of Sign Spotting instead, where the goal is to simultaneously identify and localise signs in continuous co-articulated sign videos.  ...  This work reflects only the authors view and the Commission is not responsible for any use that may be made of the information it contains.  ... 
arXiv:2210.00951v1 fatcat:ll7vmtt3djae5onercebdowr6u

2021 16TH IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021)

2021 2021 16th IEEE International Conference on Automatic Face and Gesture Recognition (FG 2021)  
AnonySign: Novel Human Appearance Synthesis for Sign Language Video Anonymisation Ben Saunders; Necati Cihan Camgoz; Richard Bowden 8.  ...  Dynamic Cross-Feature Fusion for American Sign Language Translation Tejaswini Ananthanarayana; Nikunj Kotecha; Priyanshu Srivastava; Lipisha Chaudhary; Nicholas Wilkins; Ifeoma Nwogu 56.  ... 
doi:10.1109/fg52635.2021.9667043 fatcat:q67llypbybbrbacdiyia6hs2pe

Sign Segmentation with Changepoint-Modulated Pseudo-Labelling

Katrin Renz, Nicolaj C. Stache, Neil Fox, Gul Varol, Samuel Albanie
2021 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
We make the following contributions: (1) We motivate and introduce the task of source-free domain adaptation for sign language segmentation, in which labelled source data is available for an initial training  ...  The objective of this work is to find temporal boundaries between signs in continuous sign language.  ...  The authors thank Andrew Zisserman for suggestions and Cihan Camgöz for assistance with data preparation.  ... 
doi:10.1109/cvprw53098.2021.00379 fatcat:ou62kybf6rajnfm76mczpqtozu

Cross-modal Neural Sign Language Translation

Amanda Cardoso Duarte
2019 Proceedings of the 27th ACM International Conference on Multimedia - MM '19  
With a parallel corpus of almost 60 hours of sign language videos (collected with both RGB and depth sensor data) and the corresponding speech transcripts for over 2500 instructional videos, How2Sign is  ...  Current computational approaches in this general research area have focused specifically on sign language recognition and the translation of sign language to text.  ...  To the best of our knowledge there is no dataset or study that achieved sign language translation directly from speech in a large scale and/or in a non-constrained domain.  ... 
doi:10.1145/3343031.3352587 dblp:conf/mm/Duarte19 fatcat:f5rhogpamjexpj4fyoz2kqvjxq

Pose-based Body Language Recognition for Emotion and Psychiatric Symptom Interpretation [article]

Zhengyuan Yang, Amanda Kay, Yuncheng Li, Wendi Cross, Jiebo Luo
2020 arXiv   pre-print
Inspired by the human ability to infer emotions from body language, we propose an automated framework for body language based emotion recognition starting from regular RGB videos.  ...  We first validate the accuracy and transferability of the proposed body language recognition method on several public action recognition datasets.  ...  The high recognition accuracy proves that the pose feature captures invariant information to represent actions and has an excellent ability for domain transfer even without fine-tuning.  ... 
arXiv:2011.00043v1 fatcat:silmv6y77fhsvicu3gpgb6mcvu

Learning Visual Models using a Knowledge Graph as a Trainer [article]

Sebastian Monka, Lavdim Halilaj, Stefan Schmid, Achim Rettinger
2021 arXiv   pre-print
We evaluate KG-NN on visual transfer learning tasks for classification using the mini-ImageNet dataset and its derivatives, as well as road sign recognition datasets from Germany and China.  ...  The results show that a visual model trained with a knowledge graph as a trainer outperforms a model trained with cross-entropy in all experiments, in particular when the domain gap increases.  ...  Acknowledgement This publication was created as part of the research project "KI Delta Learning" (project number: 19A19013D) funded by the Federal Ministry for Economic Affairs and Energy (BMWi) on the  ... 
arXiv:2102.08747v2 fatcat:cwcc7ric5fb35b5lwladfpckpq

Sign Language Recognition using Neural Networks

Sabaheta Djogic, Gunay Karli
2014 TEM Journal  
This encouraged us to develop a system for the Bosnian sign language since there is a need for such system.  ...  – Sign language plays a great role as communication media for people with hearing difficulties.In developed countries, systems are made for overcoming a problem in communication with deaf people.  ...  The image processing phase of sign recognition procedure obliges a lot of estimations, which presents inertness in the video stream..  ... 
doaj:052adbcd3f264a038058fc7dc73e975e fatcat:rnklez2jnbdanamprcxkvuqxya

SubUNets: End-to-End Hand Shape and Continuous Sign Language Recognition

Necati Cihan Camgoz, Simon Hadfield, Oscar Koller, Richard Bowden
2017 2017 IEEE International Conference on Computer Vision (ICCV)  
The proposed techniques are demonstrated in the challenging domain of sign language recognition.  ...  Furthermore, we are able to obtain comparable sign recognition rates to previous research, without the need for an alignment step to segment out the signs for recognition.  ...  Acknowledgement This work was funded by the SNSF Sinergia project Scalable Multimodal Sign Language Technology for Sign Language Learning and Assessment (SMILE)" grant agreement number CRSII2 160811.  ... 
doi:10.1109/iccv.2017.332 dblp:conf/iccv/CamgozHKB17 fatcat:cpew4nwob5ggbiraf724xykny4

A Comprehensive Review of Sign Language Recognition: Different Types, Modalities, and Datasets [article]

Dr. M. Madhiarasan, Prof. Partha Pratim Roy
2022 arXiv   pre-print
Sign Language Recognition (SLR) is a fascinating research area and a crucial task concerning computer vision and pattern recognition.  ...  Finally, we find the research gap and limitations in this domain and suggest future directions.  ...  [133] presented isolated sign language recognition using TK-3d convNet (transferring cross-domain knowledge-based 3D convolution network).  ... 
arXiv:2204.03328v1 fatcat:72kb7zz5xfaqxa2l5sz22drrwi

Master Thesis: Neural Sign Language Translation by Learning Tokenization [article]

Alptekin Orbay
2020 arXiv   pre-print
The tokenization part focuses on how Sign Language (SL) videos should be represented to be fed into the other part.  ...  We succeed in enabling knowledge transfer between SLs and improve translation quality by 5 points in BLEU-4 and 8 points in ROUGE scores.  ...  Furthermore, 3D-CNNs may enable information transfer from human action recognition. In this thesis, we emphasize that hand shapes are very important for sign languages.  ... 
arXiv:2011.09289v1 fatcat:44ix6kzoozhd5c6awbfa5gi7fu
« Previous Showing results 1 — 15 out of 25,953 results