… This article focuses on how using alt text, color contrast analysis, closed captioning, and
transcripts can improve the accessibility of library video tutorials, assisting the disabled and non-…

… Thus, strategies have been devised to harness synthetic captions generated by image
captioning model [34], which has augmented model performance notably [5, 14, 15, 22, 35, 51] by …

… Finally, this review paper discusses several unsolved issues in the image captioning task. …
Image captioning automatically generates textual description of contents from an image in a …

… automatically generated image captions. They compressed the 341 descriptions by considering
both linguistic fluency and con- 342 sistency of the language and visual elements. The …

… it in Image Captioning is also increased. This paper presents a literature review of image
captioning … the result of extracting alternative text associated with images from the data source …

… We show how to improve image captioning by improving the quality of the downstream
task data 23 through data curation: the process of dynamically updating the samples during …

… present in the image, their attributes, … images and audio descriptions in Portuguese. As far
as we know, this is the first dataset proposed for the Image Captioning problem with captions1 …

… image captioning models and using large-scale image-text … In this work, we explore an
alternative to web-crawled data … artificial pictures are useful for the task of image captioning, …

… automatic image captioning and grounded language understanding tasks. It comprises
31,000 images … by 158 thousand human-written captions. The dataset includes detectors for …