Browsing Documental Repository by Author "1107"
Results Per Page
ItemAudiovisual annotation in the study of physics( 2018) Marçal,J ; Borges,MM ; Carvalho,P ; Paula Viana ; 1107The support of video in the learning environment is nowadays used to many ends, either for demonstration, research or share. It is intended to reinforce the space before and after class and introduce a new dynamic and interaction in the classroom itself. Pedagogical innovation may be achieved by different approaches to motivate students and obtain better results. This paper presents a revision of the literature about the potential of using video annotation in the education context, specifically in the domain of Physics, using an open source annotation tool. The creation of audiovisual references, either for quick access to parts of organized video annotated content by the teacher, knowledge building or revision by and for other students is analyzed. This study is complemented with a testbed, showing the potential of using audiovisual annotated content, within a k-12 context. Students were invited to select video content, annotate, organize and publish the annotations, which could support the learning process in the domain of Physics. Results show that most of the aspects under analysis received a positive evaluation. The only exception relates to the capacity of the approach to motivated students to the study of Physics, as most of the students did not see this methodology as a motivating means. © 2018 ACM
ItemImproving Audiovisual Content Annotation Through a Semi-automated Process Based on Deep Learning( 2018) Paula Viana ; Maria Teresa Andrade ; Pedro Miguel Carvalho ; Vilaça,L ; 1107 ; 4358 ; 400Over the last years, Deep Learning has become one of the most popular research fields of Artificial Intelligence. Several approaches have been developed to address conventional challenges of AI. In computer vision, these methods provide the means to solve tasks like image classification, object identification and extraction of features. In this paper, some approaches to face detection and recognition are presented and analyzed, in order to identify the one with the best performance. The main objective is to automate the annotation of a large dataset and to avoid the costy and time-consuming process of content annotation. The approach follows the concept of incremental learning and a R-CNN model was implemented. Tests were conducted with the objective of detecting and recognizing one personality within image and video content. Results coming from this initial automatic process are then made available to an auxiliary tool that enables further validation of the annotations prior to uploading them to the archive. Tests show that, even with a small size dataset, the results obtained are satisfactory. © 2020, Springer Nature Switzerland AG.
ItemYouTube timed metadata enrichment using a collaborative approach( 2019) Paula Viana ; José Pedro Pinto ; 5865 ; 1107Although the growth of video content in online platforms has been happening for some time, searching and browsing these assets is still very inefficient as rich contextual data that describes the content is still not available. Furthermore, any available descriptions are, usually, not linked to timed moments of content. In this paper, we present an approach for making social web videos available on YouTube more accessible, searchable and navigable. By using the concept of crowdsourcing to collect the metadata, our proposal can contribute to easily enhance content uploaded in the YouTube platform. Metadata, collected as a collaborative annotation game, is added to the content as time-based information in the form of descriptions and captions using the YouTube API. This contributes for enriching video content and enabling navigation through temporal links. © Springer Nature Switzerland AG 2019.