This paper presents SignBuddy, the result of ongoing co-created sign language processing research. Mostsign language processing research is performed by hearing, non-signing researchers. Even though co-creation efforts have recently increased, technical research still often fails to mention if (and how) co-creation was involved in the research process. SignBuddy is a co-created research tool developed through apartnership between the Flemish Sign Language Centre, a deaf-led organisation, and Ghent University. While respecting elemental concepts of co-creation - i.e. (i) defining common goals and (ii) building a formal and sustainable relationship between users/consumers and researchers/developers and respectingthe five lessons in co-creation - the platform successfully supported the development of the first fully scalable sign-to-text dictionary search system, built into the Flemish Sign Language–Dutch onlinedictionary. SignBuddy functions as a crowdsourcing interface for in-the-wild collection of modelevaluation data, gathering example queries for quantitative performance analysis and user feedback forqualitative assessment. This human evaluation allows us to shape the application based on the end-users'needs. Addressing the need for models that support large dictionaries (over ten thousand signs), we propose a scalable one-shot sign language recognition method and achieve state-of-the-art results. Beyond the co-created application itself, this work provides insights into the co-creation process - clarifying roles, shared goals, and responsibilities - and offers conclusions to guide future co-created sign language processing research.

Isolated Sign Language Recognition (ISLR) is crucial for scalable sign language technology, yet language-specific approaches limit current models. To address this, we propose a one-shot learning approach that generalises across languages and evolving vocabularies. Our method involves pretraining a model to embed signs based on essential features and using a dense vector search for rapid, accurate recognition of unseen signs. We achieve state-of-the-art results, including 50.8 unique signs from a different language than the training set. Our approach is robust across languages and support sets, offering a scalable, adaptable solution for ISLR. Co-created with the Deaf and Hard of Hearing (DHH) community, this method aligns with real-world needs, and advances scalable sign language recognition.

Social robots should be able to detect social signals sent by their user, such as when the robot made a mistake or the user feels awkward. As our submission to the ERR@HRI challenge, we present a number of neural and traditional machine learning models to predict when this occurs in a human-robot conversation, based on facial expressions, body pose, and non-verbal speech characteristics. The small size of the dataset, imbalance in the label distribution, and low-frequency label annotations provided significant challenges. However, three of our approaches show promising results: modifying the training of a gated recurrent unit (GRU) model to predict at lower frequency than that of the input features, using an embedding layer and convolutional neural network to pre-process temporal data before feeding it to the GRU, and using traditional random forests.

Automatic translation from signed to spoken languages is an interdisciplinary research domain on the intersection of computer vision, machine translation (MT), and linguistics. While the domain is growing in terms of popularity-the majority of scientific papers on sign language (SL) translation have been published in the past five years-research in this domain is performed mostly by computer scientists in isolation. This article presents an extensive and cross-domain overview of the work on SL translation. We first give a high level introduction to SL linguistics and MT to illustrate the requirements of automatic SL translation. Then, we present a systematic literature review of the state of the art in the domain. Finally, we outline important challenges for future research. We find that significant advances have been made on the shoulders of spoken language MT research. However, current approaches often lack linguistic motivation or are not adapted to the different characteristics of SLs. We explore challenges related to the representation of SL data, the collection of datasets and the evaluation of SL translation models. We advocate for interdisciplinary research and for grounding future research in linguistic analysis of SLs. Furthermore, the inclusion of deaf and hearing end users of SL translation applications in use case identification, data collection, and evaluation, is of utmost importance in the creation of useful SL translation models.

SignON (https://signon-project.eu/) is a Horizon 2020 project, running from 2021 until the end of 2023, which addresses the lack of technology and services for the automatic translation between sign languages (SLs) and spoken languages, through an inclusive, human-centric solution, hence contributing to the repertoire of communication media for deaf, hard of hearing (DHH) and hearing individuals. In this paper, we present an update of the status of the project, describing the approaches developed to address the challenges and peculiarities of SL machine translation (SLMT).

To search for an unknown sign in a sign language dictionary, users typically indicate parameters of the query, e.g., hand shape and signing location. Recent advances in sign language recognition enable video-based sign language dictionary search. In such a system, users can record an unknown sign and retrieve a list of signs that look similar, preferably including the queried sign as one of the top results. We have realized such a system by interpreting it as a dense vector search task. First, we learn a mapping (embedding) from sign videos to a vector space. The dictionary can then be searched by looking for the vectors in this space that are closest to the vector corresponding to the query. We present a proof of concept on a subset of the Flemish Sign Language dictionary. Further research is required to scale up our method to the large vocabularies of entire dictionaries.

One of the most significant challenges to sign language recognition (SLR) today is the low resource nature of sign language datasets, with many datasets being extremely low resource. Transfer learning is therefore a promising, and likely indispensable, method of increasing recognition performance. The use of pose estimation models, which are typically trained on a large and diverse population, can also aid generalization for extremely low resource sign languages. However, research on transfer learning for pose estimation keypoints as inputs has been limited. In this work, we explore transfer learning as a means to improve SLR classification performance for the extremely low resource Irish Sign Language (ISL). We show that transfer learning on larger datasets containing secondary sign languages significantly improves performance on our target sign language, ISL. To understand these results and the attributes that make one dataset better than another for pre-training, we analyse the linguistic relationships between these datasets. We find that certain attributes of datasets are associated with better transfer learning performance. We hope that our findings will not only motivate further research into transfer learning for pose keypoint-based SLR but also act as a practical guide to researchers on choosing the most suitable datasets with which to pre-train models.

Isolated Sign Language Recognition (SLR) has mostly been applied on datasets containing signs executed slowly and clearly by a limited group of signers. In real-world scenarios, however, we are met with challenging visual conditions, coarticulated signing, small datasets, and the need for signer independent models. To tackle this difficult problem, we require a robust feature extractor to process the sign language videos. One could expect human pose estimators to be ideal candidates. However, due to a domain mismatch with their training sets and challenging poses in sign language, they lack robustness on sign language data and image-based models often still outperform keypoint-based models. Furthermore, whereas the common practice of transfer learning with image-based models yields even higher accuracy, keypoint-based models are typically trained from scratch on every SLR dataset. These factors limit their usefulness for SLR. From the existing literature, it is also not clear which, if any, pose estimator performs best for SLR. We compare the three most popular pose estimators for SLR: OpenPose, MMPose and MediaPipe. We show that through keypoint normalization, missing keypoint imputation, and learning a pose embedding, we can obtain significantly better results and enable transfer learning. We show that keypoint-based embeddings contain cross-lingual features: they can transfer between sign languages and achieve competitive performance even when fine-tuning only the classifier layer of an SLR model on a target sign language. We furthermore achieve better performance using fine-tuned transferred embeddings than models trained only on the target sign language. The embeddings can also be learned in a multilingual fashion. The application of these embeddings could prove particularly useful for low resource sign languages in the future.

We consider neural sign language translation: machine translation from signed to written languages using encoder–decoder neural networks. Translating sign language videos to written language text is especially complex because of the difference in modality between source and target language and, consequently, the required video processing. At the same time, sign languages are low-resource languages, their datasets dwarfed by those available for written languages. Recent advances in written language processing and success stories of transfer learning raise the question of how pretrained written language models can be leveraged to improve sign language translation. We apply the Frozen Pretrained Transformer (FPT) technique to initialize the encoder, decoder, or both, of a sign language translation model with parts of a pretrained written language model. We observe that the attention patterns transfer in zero-shot to the different modality and, in some experiments, we obtain higher scores (from 18.85 to 21.39 BLEU-4). Especially when gloss annotations are unavailable, FPTs can increase performance on unseen data. However, current models appear to be limited primarily by data quality and only then by data quantity, limiting potential gains with FPTs. Therefore, in further research, we will focus on improving the representations used as inputs to translation models.

The SignON project (www.signon-project.eu) focuses on the research and development of a Sign Language (SL) translation mobile application and an open communications framework. SignON rectifies the lack of technology and services for the automatic translation between signed and spoken languages, through an inclusive, humancentric solution which facilitates communication between deaf, hard of hearing (DHH) and hearing individuals. We present an overview of the current status of the project, describing the milestones reached to date and the approaches that are being developed to address the challenges and peculiarities of Sign Language Machine Translation (SLMT).

Sign Languages (SLs) are the primary means of communication for at least half a million people in Europe alone. However, the development of SL recognition and translation tools is slowed down by a series of obstacles concerning resource scarcity and standardization issues in the available data. The former challenge relates to the volume of data available for machine learning as well as the time required to collect and process new data. The latter obstacle is linked to the variety of the data, i.e., annotation formats are not unified and vary amongst different resources. The available data formats are often not suitable for machine learning, obstructing the provision of automatic tools based on neural models. In the present paper, we give an overview of these challenges by comparing various SL corpora and SL machine learning datasets. Furthermore, we propose a framework to address the lack of standardization at format level, unify the available resources and facilitate SL research for different languages. Our framework takes ELAN files as inputs and returns textual and visual data ready to train SL recognition and translation models. We present a proof of concept, training neural translation models on the data produced by the proposed framework.

One of the major challenges in sign language translation from a sign language to a spoken language is the lack of parallel corpora. Recent works have achieved promising results on the RWTH-PHOENIX-Weather 2014T dataset, which consists of over eight thousand parallel sentences between German sign language and German. However, from the perspective of neural machine translation, this is still a tiny dataset. To improve the performance of models trained on small datasets, transfer learning can be used. While this has been previously applied in sign language translation for feature extraction, to the best of our knowledge, pretrained language models have not yet been investigated. We use pretrained BERT-base and mBART-50 models to initialize our sign language video to spoken language text translation model. To mitigate overfitting, we apply the frozen pretrained transformer technique: we freeze the majority of parameters during training. Using a pretrained BERT model, we outperform a baseline trained from scratch by 1 to 2 BLEU-4. Our results show that pretrained language models can be used to improve sign language translation performance and that the self-attention patterns in BERT transfer in zero-shot to the encoder and decoder of sign language translation models.