Please use this identifier to cite or link to this item:
https://www.um.edu.mt/library/oar/handle/123456789/104590| Title: | Pre-gen metrics : predicting caption quality metrics without generating captions |
| Other Titles: | Computer Vision – ECCV 2018 Workshops |
| Authors: | Tanti, Marc Gatt, Albert Muscat, Adrian |
| Keywords: | Subtitles (Motion pictures, television, etc.) Neural networks (Computer science) Versification |
| Issue Date: | 2018 |
| Publisher: | Springer International Publishing |
| Citation: | Tanti, M., Gatt, A., & Muscat, A. (2018). Pre-gen metrics : predicting caption quality metrics without generating captions. In Computer Vision – ECCV 2018 Workshops (pp. 114-123). Manhattan: Springer International Publishing. |
| Abstract: | Image caption generation systems are typically evaluated against reference outputs. We show that it is possible to predict output quality without generating the captions, based on the probability assigned by the neural model to the reference captions. Such pre-gen metrics are strongly correlated to standard evaluation metrics. |
| URI: | https://www.um.edu.mt/library/oar/handle/123456789/104590 |
| Appears in Collections: | Scholarly Works - InsLin |
Files in This Item:
| File | Description | Size | Format | |
|---|---|---|---|---|
| Pre_gen_metrics_predicting_caption_quality_metrics_without_generating_captions_2018.pdf Restricted Access | 559.26 kB | Adobe PDF | View/Open Request a copy |
Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.