Subtitles are crucial for video content understanding. However, a large amount of videos have only burned-in, hardcoded subtitles that prevent re-editing, translation, etc. In this paper, we construct deep-learning-based system the inverse conversion burned-in subtitle to file and an inpainted video, by coupling three deep neural networks (CTPN, CRNN, EdgeConnect). We evaluated performance prop...

Audiovisual material accompanied by interlingual subtitles is a powerful pedagogical tool which can help improve the vocabulary learning of second-language learners. This study was intended to determine whether or not the mode (standard and reversed) of subtitling affects the incidental vocabulary acquisition of Iranian L2 learners while watching TV programs. Forty-five participants were random...

In the last two decades, Audiovisual Translation (AVT) studies have become of interest for Second Language Acquisition (SLA) researchers, particularly regarding use subtitles in language learning activities. This paper will present an experiment aimed at investigating role subtitled 'input enhancement' SLA. The study involved a group Italian native students English as Foreign (EFL) from Milan U...

This project explores how deaf children read subtitles on television. The participants – recruited from years 3 to 6 of a mainstream school with a hearing impairment unit – were exposed to both broadcast and enhanced subtitles and their performances were compared. In particular, the focus is on identifying enhancements that can help children to understand subtitle content and to recognise new o...

We assess the performance of generic text summarization algorithms applied to films and documentaries, using the well–known behavior of summarization of news articles as reference. We use three datasets: (i) news articles, (ii) film scripts and subtitles, and (iii) documentary subtitles. Standard ROUGE metrics are used for comparing generated summaries against news abstracts, plot summaries, an...

Subband decomposition techniques have been extensively used for data coding and analysis. In most lter banks, the goal is to obtain subsampled signals corresponding to di erent spectral bands of the original data. However, this approach leads to various artifacts in images containing text, subtitles, or sharp edges. In this paper, adaptive lter banks with perfect reconstruction property are pre...

This study was an attempt to identify the interlingual strategies employed to translate English subtitles into Persian and to determine their frequency, as well. Contrary to many countries, subtitling is a new field in Iran. The study, a corpus-based, comparative, descriptive, non-judgmental analysis of an English-Persian parallel corpus, comprised English audio scripts of five movies of differ...

Subtitling for the deaf and hard-of-hearing (SDH) is an area which deserves a special attention as it ena- bles these people to access to the part of the ‘world’ intended for hearing people, including the world of ‘motion pictures’, and particularly movie sounds. Compared to linguistic sounds, non-linguistic sounds have received little attention in the field of translation, although they are in...