Russian Sign Language Dactyl Recognition
In this paper, we compare several real-time sign language dactyl recognition systems and present a new model based on deep convolutional neural networks. These systems are able to recognize Russian alphabet letters presented as static signs in Russian Sign language used by people from deaf community. In such an approach, we recognize words from Russian natural language presented by consequent hand gestures of each letter. We evaluate our approach on Russian (RSL) sign language, for which we collect our own dataset and evaluate dactyl recognition.
In this paper we focus on the problem of multi-label image recognition for visually-aware recommender systems. We propose a two stage approach in which a deep convolutional neural network is firstly fine-tuned on a part of the training set. Secondly, an attention-based aggregation network is trained to compute the weighted average of visual features in an input image set. Our approach is implemented as a mobile fashion recommender system application. It is experimentally show on the Amazon Fashion dataset that our approach achieves an F1-measure of 0.58 for 15 recommendations, which is twice as good as the 0.25 F1-measure for conventional averaging of feature vectors.
The article proposes an analysis of three semantic fields in Russian Sign Language (RSL): ‘thick’, ‘thin’ and ‘pointed’. These fields are covered in RSL with a particular group of signs, namely, size and shape specifiers (SASSes). The paper describes features of SASSes in other sign languages, known from previous research, and proposes an analysis of these signs in RSL based on a detailed study of their contexts. Particularly, the article argues for distinguishing two types of components in these signs (specified and non-specified ones), discusses the semantics of non-manual markers and describes two morphological forms of SASSes.
The authors consider the problem of human pose estimation using probabilistic convolutional neural networks. They explore ways to improve human pose estimation accuracy on standard pose estimation benchmarks MPII human pose and Leeds Sports Pose (LSP) datasets using frameworks for probabilistic deep learning. Such frameworks transform deterministic neural network into a probabilistic one and allow sampling of independent and equiprobable hypotheses (different outputs) for a given input. Overlapping body parts and body joints hidden under clothes or other obstacles make the problem of human pose estimation ambiguous. In this context to get accurate estimation of joints’ position they use uncertainty in network's predictions, which is represented by variance of hypotheses, provided by a probabilistic convolutional neural network, and confidence is characterised by mean of them. Their work is based on current CNN cascades for pose estimation. They propose and evaluate three probabilistic convolutional neural networks built on top of deterministic ones with two probabilistic deep learning frameworks – DISCO networks and Bayesian SegNet. The authors evaluate their models on standard pose estimation benchmarks and show that proposed probabilistic models outperform base deterministic ones.