NECATI CIHAN CAMGÖZ

In natural language processing (NLP) of spoken languages , word embeddings have been shown to be a useful method to encode the meaning of words. Sign languages are visual languages, which require sign embeddings to capture the visual and linguistic semantics of sign.



Unlike many common approaches to Sign Recognition, we focus on explicitly creating sign embeddings that bridge the gap between sign language and spoken language. We propose a learning framework to derive LCC (Learnt Con-trastive Concept) embeddings for sign language, a weakly supervised contrastive approach to learning sign embed-dings. We train a vocabulary of embeddings that are based on the linguistic labels for sign video. Additionally, we develop a conceptual similarity loss which is able to utilise word embeddings from NLP methods to create sign embed-dings that have better sign language to spoken language correspondence. These learnt representations allow the model to automatically localise the sign in time.



Our approach achieves state-of-the-art keypoint-based sign recognition performance on the WLASL and BOBSL datasets.

Sign Language Translation (SLT) is a challenging task that aims to generate spoken language sentences from sign language videos, both of which have different grammar and word/gloss order. From a Neural Machine Translation (NMT) perspective, the straightforward way of training translation models is to use sign language phrase-spoken language sentence pairs. However, human interpreters heavily rely on the context to understand the conveyed information , especially for sign language interpretation, where the vocabulary size may be significantly smaller than their spoken language equivalent.



Taking direct inspiration from how humans translate, we propose a novel multi-modal transformer architecture that tackles the translation task in a context-aware manner, as a human would. We use the context from previous sequences and confident predictions to disambiguate weaker visual cues. To achieve this we use complementary transformer encoders, namely: (1) A Video Encoder, that captures the low-level video features at the frame-level, (2) A Spotting Encoder, that models the recognized sign glosses in the video, and (3) A Context Encoder, which captures the context of the preceding sign sequences. We combine the information coming from these encoders in a final transformer decoder to generate spoken language translations.



We evaluate our approach on the recently published large-scale BOBSL dataset, which contains ∼1.2M sequences , and on the SRF dataset, which was part of the WMT-SLT 2022 challenge. We report significant improvements on state-of-the-art translation performance using contextual information, nearly doubling the reported BLEU-4 scores of baseline approaches.

It is common practice to represent spoken languages at their phonetic level. However, for sign languages, this implies breaking motion into its constituent motion primitives. Avatar based Sign Language Production (SLP) has traditionally done just this, building up animation from sequences of hand motions, shapes and facial expressions. However, more recent deep learning based solutions to SLP have tackled the problem using a single network that estimates the full skeletal structure.
We propose splitting the SLP task into two distinct jointly-trained sub-tasks. The first translation sub-task translates from spoken language to a latent sign language representation, with gloss supervision. Subsequently, the animation sub-task aims to produce expressive sign language sequences that closely resemble the learnt spatio-temporal representation. Using a progressive transformer for the translation sub-task, we propose a novel Mixture of Motion Primitives (MOMP) architecture for sign language animation. A set of distinct motion primitives are learnt during training, that can be temporally combined at inference to animate continuous sign language sequences.
We evaluate on the challenging RWTH-PHOENIX-Weather-2014T (PHOENIX14T) dataset, presenting extensive ablation studies and showing that MOMP outperforms baselines in user evaluations. We achieve state-of-the-art back translation performance with an 11% improvement over competing results. Importantly, and for the first time, we showcase stronger performance for a full translation pipeline going from spoken language to sign, than from gloss to sign.

Sign languages are visual languages, with vocabularies as rich as their spoken language counterparts. However, current deep-learning based Sign Language Production (SLP) models produce under-articulated skeleton pose sequences from constrained vocabularies and this limits applicability. To be understandable and accepted by the deaf, an automatic SLP system must be able to generate co-articulated photo-realistic signing sequences for large domains of discourse.
In this work, we tackle large-scale SLP by learning to co-articulate between dictionary signs, a method capable of producing smooth signing while scaling to unconstrained domains of discourse. To learn sign co-articulation, we propose a novel Frame Selection Network ( FS- NET) that improves the temporal alignment of interpolated dictionary signs to continuous signing sequences. Additionally, we propose SIGNGAN, a pose-conditioned human synthesis model that produces photo-realistic sign language videos direct from skeleton pose. We propose a novel keypoint- based loss function which improves the quality of synthesized hand images.
We evaluate our SLP model on the large-scale meineDGS (mDGS) corpus, conducting extensive user evaluation showing our FS-NET approach improves coarticulation of interpolated dictionary signs. Additionally, we show that SIGNGAN significantly outperforms all baseline methods for quantitative metrics, human perceptual studies and native deaf signer comprehension.

Sign languages are visual languages produced by the movement of the hands, face, and body. In this paper, we evaluate representations based on skeleton poses, as these are explainable, person-independent, privacy-preserving, low-dimensional representations. Basically, skeletal representations generalize over an individual's appearance and background, allowing us to focus on the recognition of motion. But how much information is lost by the skeletal representation? We perform two independent studies using two state-of-the-art pose estimation systems. We analyze the applicability of the pose estimation systems to sign language recognition by evaluating the failure cases of the recognition models. Importantly, this allows us to characterize the current limitations of skeletal pose estimation approaches in sign language recognition.

Disentangled representations support a range of downstream tasks including causal reasoning, generative modeling, and fair machine learning. Unfortunately, disentanglement has been shown to be impossible without the incorporation of supervision or inductive bias. Given that supervision is often expensive or infeasible to acquire, we choose to incorporate structural inductive bias and present an unsupervised, deep State-Space-Model for Video Disentanglement (VDSM). The model disentangles latent time-varying and dynamic factors via the incorporation of hierarchical structure with a dynamic prior and a Mixture of Experts decoder. VDSM learns separate disentangled representations for the identity of the object or person in the video, and for the action being performed. We evaluate VDSM across a range of qualitative and quantitative tasks including identity and dynamics transfer, sequence generation, Fréchet Inception Distance, and factor classification. VDSM achieves state-of-the-art performance and exceeds adversarial methods, even when the methods use additional supervision.

Sign language recognition (SLR) involves identifying the form and meaning of isolated signs or sequences of signs. To our knowledge, the combination of SLR and sign language assessment is novel. The goal of an ongoing three-year project in Switzerland is to pioneer an assessment system for lexical signs of Swiss German Sign Language (Deutschschweizerische Gebardensprache, DSGS) that relies on SLR. The assessment system aims to give adult L2 learners of DSGS feedback on the correctness of the manual parameters (handshape, hand position, location, and movement) of isolated signs they produce. In its initial version, the system will include automatic feedback for a subset of a DSGS vocabulary production test consisting of 100 lexical items. To provide the SLR component of the assessment system with sufficient training samples, a large-scale dataset containing videotaped repeated productions of the 100 items of the vocabulary test with associated transcriptions and annotations was created, consisting of data from 11 adult L1 signers and 19 adult L2 learners of DSGS. This paper introduces the dataset, which will be made available to the research community.

We propose a novel deep learning approach to solve simultaneous alignment and recognition problems (referred to as "Sequence-to-sequence" learning). We decompose the problem into a series of specialised expert systems referred to as SubUNets. The spatio-temporal relationships between these SubUNets are then modelled to solve the task, while remaining trainable end-to-end. The approach mimics human learning and educational techniques, and has a number of significant advantages. SubUNets allow us to inject domain-specific expert knowledge into the system regarding suitable intermediate representations. They also allow us to implicitly perform transfer learning between different interrelated tasks, which also allows us to exploit a wider range of more varied data sources. In our experiments we demonstrate that each of these properties serves to significantly improve the performance of the overarching recognition system, by better constraining the learning problem. The proposed techniques are demonstrated in the challenging domain of sign language recognition. We demonstrate state-of-the-art performance on hand-shape recognition (outperforming previous techniques by more than 30%). 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.

In this paper, we propose a novel particle filter based probabilistic forced alignment approach for training spatio-temporal deep neural networks using weak border level annotations. The proposed method jointly learns to localize and recognize isolated instances in continuous streams. This is done by drawing training volumes from a prior distribution of likely regions and training a discriminative 3D-CNN from this data. The classifier is then used to calculate the posterior distribution by scoring the training examples and using this as the prior for the next sampling stage. We apply the proposed approach to the challenging task of large-scale user-independent continuous gesture recognition. We evaluate the performance on the popular ChaLearn 2016 Continuous Gesture Recognition (ConGD) dataset. Our method surpasses state-of-the-art results by obtaining 0.3646 and 0.3744 Mean Jaccard Index Score on the validation and test sets of ConGD, respectively. Furthermore, we participated in the ChaLearn 2017 Continuous Gesture Recognition Challenge and was ranked 3rd. It should be noted that our method is learner independent, it can be easily combined with other approaches.

In this paper, we propose using 3D Convolutional Neural Networks for large scale user-independent continuous gesture recognition. We have trained an end-to-end deep network for continuous gesture recognition (jointly learning both the feature representation and the classifier). The network performs three-dimensional (i.e. space-time) convolutions to extract features related to both the appearance and motion from volumes of color frames. Space-time invariance of the extracted features is encoded via pooling layers. The earlier stages of the network are partially initialized using the work of Tran et al. before being adapted to the task of gesture recognition. An earlier version of the proposed method, which was trained for 11,250 iterations, was submitted to ChaLearn 2016 Continuous Gesture Recognition Challenge and ranked 2nd with the Mean Jaccard Index Score of 0.269235. When the proposed method was further trained for 28,750 iterations, it achieved state-of-the-art performance on the same dataset, yielding a 0.314779 Mean Jaccard Index Score.