Visual Content Recognition by Exploiting Semantic Feature Map with Attention and Multi-task Learning

Author(s):  
Rui-Wei Zhao ◽  
Qi Zhang ◽  
Zuxuan Wu ◽  
Jianguo Li ◽  
Yu-Gang Jiang
2021 ◽  
Vol 33 (3) ◽  
pp. 363-375
Author(s):  
Fan Guo ◽  
Weiqing Li ◽  
Xin Zhao ◽  
Beiji Zou

Sensors ◽  
2021 ◽  
Vol 21 (16) ◽  
pp. 5476
Author(s):  
Rui Wang ◽  
Jialing Zou ◽  
James Zhiqing Wen

Monocular depth estimation based on unsupervised learning has attracted great attention due to the rising demand for lightweight monocular vision sensors. Inspired by multi-task learning, semantic information has been used to improve the monocular depth estimation models. However, multi-task learning is still limited by multi-type annotations. As far as we know, there are scarcely any large public datasets that provide all the necessary information. Therefore, we propose a novel network architecture Semantic-Feature-Aided Monocular Depth Estimation Network (SFA-MDEN) to extract multi-resolution depth features and semantic features, which are merged and fed into the decoder, with the goal of predicting depth with the support of semantics. Instead of using loss functions to relate the semantics and depth, the fusion of feature maps for semantics and depth is employed to predict the monocular depth. Therefore, two accessible datasets with similar topics for depth estimation and semantic segmentation can meet the requirements of SFA-MDEN for training sets. We explored the performance of the proposed SFA-MDEN with experiments on different datasets, including KITTI, Make3D, and our own dataset BHDE-v1. The experimental results demonstrate that SFA-MDEN achieves competitive accuracy and generalization capacity compared to state-of-the-art methods.


2019 ◽  
Vol 62 (12) ◽  
pp. 4464-4482 ◽  
Author(s):  
Diane L. Kendall ◽  
Megan Oelke Moldestad ◽  
Wesley Allen ◽  
Janaki Torrence ◽  
Stephen E. Nadeau

Purpose The ultimate goal of anomia treatment should be to achieve gains in exemplars trained in the therapy session, as well as generalization to untrained exemplars and contexts. The purpose of this study was to test the efficacy of phonomotor treatment, a treatment focusing on enhancement of phonological sequence knowledge, against semantic feature analysis (SFA), a lexical-semantic therapy that focuses on enhancement of semantic knowledge and is well known and commonly used to treat anomia in aphasia. Method In a between-groups randomized controlled trial, 58 persons with aphasia characterized by anomia and phonological dysfunction were randomized to receive 56–60 hr of intensively delivered treatment over 6 weeks with testing pretreatment, posttreatment, and 3 months posttreatment termination. Results There was no significant between-groups difference on the primary outcome measure (untrained nouns phonologically and semantically unrelated to each treatment) at 3 months posttreatment. Significant within-group immediately posttreatment acquisition effects for confrontation naming and response latency were observed for both groups. Treatment-specific generalization effects for confrontation naming were observed for both groups immediately and 3 months posttreatment; a significant decrease in response latency was observed at both time points for the SFA group only. Finally, significant within-group differences on the Comprehensive Aphasia Test–Disability Questionnaire ( Swinburn, Porter, & Howard, 2004 ) were observed both immediately and 3 months posttreatment for the SFA group, and significant within-group differences on the Functional Outcome Questionnaire ( Glueckauf et al., 2003 ) were found for both treatment groups 3 months posttreatment. Discussion Our results are consistent with those of prior studies that have shown that SFA treatment and phonomotor treatment generalize to untrained words that share features (semantic or phonological sequence, respectively) with the training set. However, they show that there is no significant generalization to untrained words that do not share semantic features or phonological sequence features.


2019 ◽  
Vol 5 (5) ◽  
Author(s):  
Matthew E. Ferrandino ◽  
Brad Osborn

Music video combines moving images with a preexisting song. The narrative implied by a music video’s visual content can either support or seem at odds with the narrative suggested by the song’s music and lyrics, in ways that have fascinating repercussions. In this video, we explore four different relationships between image and sound and how these interactions influence our interpretation of music video.


Author(s):  
Tamara Vázquez-Barrio ◽  
Teresa Torrecillas-Lacave ◽  
Rebeca Suárez-Álvarez

Traditional television coexists with formats that originated on the internet, as well as on-demand consumption, paid television, and other audio-visual content distribution platforms. Audience data reveal a steady decline in television viewership, and digital technologies now allow any citizen to produce audio-visual content and distribute it for mass consumption through the internet. Given this new audio-visual ecosystem, the aim of this research is to ascertain whether there are any signs of a crisis regarding the dominance of television as a means of disseminating the products of the culture industry. Disinterest or indifference toward conventional programming by users would reveal a danger to the broadcast industry. In contrast, the consumption of television products through other channels would imply the retention of television audiences through the internet. This study analyzes perceptions regarding television through five online discussion groups. Three conclusions can be drawn: Firstly, television holds a prominent place in the daily lives of those who use it, including the youngest participants, despite the fact that audiences have declined in recent years. The second conclusion states that the perception of television is positive and associated with disengagement, relaxation, and family gatherings, which can be combined with individual consumption at other times of the day. As a third conclusion, this study reveals the high degree of compatibility between the internet and television screens, as new forms of consumption are emerging, yet there is still a predominant interest in content on television and from the mass culture industry. Resumen La televisión tradicional convive con formatos originados en internet, con el consumo bajo demanda, con la televisión de pago y con otras plataformas de distribución de contenido audiovisual. Los datos de audiencias muestran un descenso continuado de telespectadores y las tecnologías digitales permiten a cualquier ciudadano producir contenidos audiovisuales y distribuirlos para el consumo masivo a través de la Red. Ante este nuevo ecosistema audiovisual, el objetivo de esta investigación es comprobar si se pueden advertir signos de una crisis de la supremacía del televisor como medio de difusión de industria cultural. El desapego o indiferencia de los usuarios hacia la programación convencional evidenciaría un peligro para la televisión. Al contrario, el consumo de productos televisivos a través de otras pantallas implicaría el mantenimiento de las audiencias televisivas a través de internet. La investigación analiza las percepciones sobre la televisión mediante cinco grupos de discusión online. Se extraen tres conclusiones. La primera, que la televisión ocupa una posición relevante en la cotidianeidad de los participantes, incluidos los más jóvenes, a pesar de que las audiencias han descendido en los últimos años. La segunda, que la percepción sobre la televisión es positiva y se asocia a la desconexión, el relax y a un momento de reunión familiar compaginable con consumos individualizados en otros momentos del día. Tercera, el estudio demuestra el alto grado de compatibilidad entre internet y la pantalla del televisor porque surgen nuevas formas de consumo, pero se mantiene un interés predominante por los contenidos televisivos y de la gran industria cultural.


2013 ◽  
Author(s):  
Peter S. Schaefer ◽  
Clinton R. Irvin ◽  
Paul N. Blankenbeckler ◽  
C. J. Brogdon
Keyword(s):  

Sign in / Sign up

Export Citation Format

Share Document