ViTag: Automatic Video Tagging Using Segmentation and Conceptual Inference

Patwardhan, Abhishek A; Das, Santanu; Varshney, Sakshi; Desarkar, Maunendra Sankar; Dogra, Debi Prosad

doi:10.1109/bigmm.2019.00-12

Cited by 8 publications

(2 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There are diverse methods used by researchers for image and video tagging. Most of the methods used a supervised approach, as in [4] [5] [7] [12] [14] [16] [19] [20], whereas few researchers adopted an unsupervised approach, as in [6] [8] [15] [25]. Both the supervised and unsupervised approaches used for image and video tagging leveraged different types of features.…”

Section: Discussionmentioning

confidence: 99%

Automatic Image and Video Tagging Survey

Athab,

Abdullah Karim

2023

Iraqi Journal of Science

View full text Add to dashboard Cite

Marking content with descriptive terms that depict the image content is called “tagging,” which is a well-known method to organize content for future navigation, filtering, or searching. Manually tagging video or image content is a time-consuming and expensive process. Accordingly, the tags supplied by humans are often noisy, incomplete, subjective, and inadequate. Automatic Image Tagging can spontaneously assign semantic keywords according to the visual information of images, thereby allowing images to be retrieved, organized, and managed by tag. This paper presents a survey and analysis of the state-of-the-art approaches for the automatic tagging of video and image data. The analysis in this paper covered the publications on tagging in Scopus and the Web of Science databases from 2008 to 2022.

show abstract

Section: Discussionmentioning

confidence: 99%

Automatic Image and Video Tagging Survey

Athab,

Abdullah Karim

2023

Iraqi Journal of Science

View full text Add to dashboard Cite

show abstract

“…To the best of our knowledge, no state-of-the-art models suggest both tags and thumbnails relying on a sole algorithm; • (ii) as the tags are suggested considering the latest relevant popular cultural heritage topics, unlike most of the state-of-the-art tools (e.g., the proposal by Jin et al [11]), our proposal can constantly revitalize the associated tags; • (iii) the option, for the final user, of setting a trade-off between quality and quantity of suggested items. Unlike the current tools (e.g., the work of Patwardhan et al [19]), the final user can decide if it would be worth enriching a gem with many tags/thumbnails (with the risk of assigning less relevant items) or preferring to select only relevant tags/thumbnails having a confidence score higher than a certain threshold, assuming the risk of not retrieving any items (if there are no items with the confidence score higher than the fixed threshold).…”

Section: Authors and Reference Title Main Topicmentioning

confidence: 99%