Search

Semantic similarity between texts can be defined based on their meaning. Assessing the textual similarity is a prerequisite in almost all applications in the field of language processing and information retrieval. However, the diversity in the sentence structure makes it formidable to estimate the similarity. Some sentences pairs are lexicographically similar but semantically dissimilar. That is why the trivial lexical overlapping is not enough for measuring the similarity. To attain the semanticity of sentences, the context of the words and the structure of the sentence should be considered. In this paper, we propose a new method for capturing the semantic similarity between sentences based on their grammatical roles through word semantics. First, the sentences are divided grammatically into different parts where each part is considered as a grammatical role. Then multiple new measures are introduced to estimate the role-based similarity exploiting word semantics considering the sentence structure. The proposed similarity measures focus on inter-role and intra-role similarity between the sentence-pair. The word-level semantic information is extracted from a pre-trained word-embedding model. The performance of the proposed method was verified by conducting a wide range of experiments on the SemEval STS dataset. The experimental results indicated the effectiveness of the proposed method in terms of different standard evaluation metrics and outperformed some known related works.

Current action recognition studies enjoy the benefits of two neural network branches, spatial and temporal. This work aims to extend the previous work by introducing a fusion of spatial and temporal branches to provide superior action recognition capability toward multi-label multi-class classification problems. In this paper, we propose three fusion models with different fusion strategies. We first build several efficient temporal Gaussian mixture (TGM) layers to form spatial and temporal branches to learn a set of features. In addition to these branches, we introduce a new deep spatio-temporal branch consisting of a series of TGM layers to learn the features that emerged from the existing branches. Each branch produces a temporal-aware feature that assists the model in understanding the underlying action in a video. To verify the performance of our proposed models, we performed extensive experiments using the well-known MultiTHUMOS benchmarking dataset. The results demonstrate the importance of our proposed deep fusion mechanism, contributing to the overall score while keeping the number of parameters small.

Spatiotemporal and motion feature representations are the key to video action recognition. Typical previous approaches are to utilize 3D CNNs to cope with both spatial and temporal features, but they suffer from huge computations. Other approaches are to utilize (1+2)D CNNs to learn spatial and temporal features in an efficient way, but they neglect the importance of motion representations. To overcome problems with previous approaches, we propose a novel block which makes it possible to alleviate the aforementioned problems, since our block can capture spatial and temporal features more faithfully and efficiently learn motion features. This proposed block includes Motion Excitation (ME), Multi-view Excitation (MvE), and Densely Connected Temporal Aggregation (DCTA). The purpose of ME is to encode feature-level frame differences; MvE is designed to enrich spatiotemporal features with multiple view representations adaptively; and DCTA is to model long-range temporal dependencies. We inject the proposed building block, which we refer to as the META block (or simply “META”), into 2D ResNet-50. Through extensive experiments, we demonstrate that our proposed method architecture outperforms previous CNN-based methods in terms of “Val Top-1 %” measure with Something-Something v1 and Jester datasets, while the META yielded competitive results with the Moment-in-Time Mini dataset.