Semantic features and high-order physical features fusion for action recognition

2021 ◽  
Author(s):  
Limin Xia ◽  
Wentao Ma ◽  
Lu Feng
2021 ◽  
Vol 45 (4) ◽  
pp. 351-363
Author(s):  
Goranka Blagus Bartolec ◽  
Ivana Matas Ivanković

Proverbs as concise textual structures are primarily defined as oral (folk) literary forms in which universal thoughts are expressed on the basis of individual experiences understandable to speakers of the language, i.e., of the social community in which they originated. In relation to, for example, idioms, the use of proverbs in today’s public discourse is much rarer, and proverbs in Croatian are most often recorded in printed form, while online edited lexicographic sources of proverbs are rare. Folk customs, human character and physical features, social and religious values, the relation of human and nature are the most common motives in proverbs. Male-female relationships are also the subject of numerous proverbs. Given the past times when they were created, they can be considered the source of a stereotypical image of the status of women and men in society that exists in human consciousness. Based on proverbs with the component woman, grandmother, mother, daughter, sister, girlfriend, widow, father, son, husband…, this paper will analyze proverbs with the topic of male-female relations, e.g. Ljubav daj ženi, ali tajnu odaj samo majci i sestri. (Give your love to your wife, but reveal the secret only to your mother and sister.), or proverbs referring to an individual feature attributed to a man or a woman, e.g., Kakvo drvo, takav klin, kakav otac takav sin. (Like tree, like wedge; like father, like son.)., Ženi sina kad hoćeš, a kćer kad možeš. (Marry a son when you want and a daughter when you can.). The analysis includes the following: 1. representation of proverbs in other lexicographic (printed and online sources), 2. representation of such proverbs in contemporary public discourse, 3. structural and semantic features of proverbs motivated by male-female relationships. In conclusion, the role of proverbs on the topic of male and female in the contemporary context is discussed – what is their perspective and whether the corpus has replaced traditional recorders and word of mouth today.


2007 ◽  
Vol 18 (10) ◽  
pp. 845-849 ◽  
Author(s):  
Mika Koivisto ◽  
Antti Revonsuo

Inattentional blindness refers to the failure to see an unexpected object that one may be looking at directly when one's attention is elsewhere. We studied whether a stimulus whose meaning is relevant to the attentional goals of the observer will capture attention and escape inattentional blindness. The results showed that an unexpected stimulus belonging to the attended semantic category but not sharing physical features with the attended stimuli was detected more often than a semantically unrelated stimulus. This effect was larger when the unexpected stimuli were words than when they were pictures. The results imply that the semantic relation between the observer's attentional set and the unexpected stimulus plays a crucial role in inattentional blindness: An unexpected stimulus semantically related to the observer's current interests is likely to be seen, whereas unrelated unexpected stimuli are unseen. Attentional selection may thus be driven by purely semantic features: Meaning may determine whether or not one sees a stimulus.


Author(s):  
Yang Zhou ◽  
Bingbing Ni ◽  
Shuicheng Yan ◽  
Pierre Moulin ◽  
Qi Tian

2016 ◽  
Vol 6 (10) ◽  
pp. 2002
Author(s):  
Chengyu Nan

Typologically, English, Chinese and Korean belong to three different types of language. English is inflectional, Chinese is isolating and Korean is agglutinative. Therefore, words of perception in these three languages show some different semantic features. But due to similar physical features and physiological phenomenon, people speaking English, Chinese or Korean language use the same word of perception to express the same meaning or feeling. This paper makes a comparative case study of mouth, 嘴/口 and입, which have rich polysemous features. Their meanings are extended from “the part of human body” to the concrete “entrance” or “person” and then to the abstract “speech act” or “way of speaking”. The meaning extension shows semantic symmetry and asymmetry both interlingually and intralingually in the expressions not only with mouth, 嘴/口 and입 and other words of perception in three languages.


Symmetry ◽  
2020 ◽  
Vol 12 (10) ◽  
pp. 1589
Author(s):  
Zeyuan Hu ◽  
Eung-Joo Lee

Traditional convolution neural networks have achieved great success in human action recognition. However, it is challenging to establish effective associations between different human bone nodes to capture detailed information. In this paper, we propose a dual attention-guided multiscale dynamic aggregate graph convolution neural network (DAG-GCN) for skeleton-based human action recognition. Our goal is to explore the best correlation and determine high-level semantic features. First, a multiscale dynamic aggregate GCN module is used to capture important semantic information and to establish dependence relationships for different bone nodes. Second, the higher level semantic feature is further refined, and the semantic relevance is emphasized through a dual attention guidance module. In addition, we exploit the relationship of joints hierarchically and the spatial temporal correlations through two modules. Experiments with the DAG-GCN method result in good performance on the NTU-60-RGB+D and NTU-120-RGB+D datasets. The accuracy is 95.76% and 90.01%, respectively, for the cross (X)-View and X-Subon the NTU60dataset.


2012 ◽  
Vol 116 (3) ◽  
pp. 361-377 ◽  
Author(s):  
Jingen Liu ◽  
Yang Yang ◽  
Imran Saleemi ◽  
Mubarak Shah

Sign in / Sign up

Export Citation Format

Share Document