Egoshots, an ego-vision life-logging dataset and semantic fidelity metric to evaluate diversity in image captioning models - Inria - Institut national de recherche en sciences et technologies du numérique Accéder directement au contenu
Communication Dans Un Congrès Année : 2020

Egoshots, an ego-vision life-logging dataset and semantic fidelity metric to evaluate diversity in image captioning models

Résumé

Image captioning models have been able to generate grammatically correct and human understandable sentences. However most of the captions convey limited information as the model used is trained on datasets that do not caption all possible objects existing in everyday life. Due to this lack of prior information most of the captions are biased to only a few objects present in the scene, hence limiting their usage in daily life. In this paper, we attempt to show the biased nature of the currently existing image captioning models and present a new image captioning dataset, Egoshots, consisting of 978 real life images with no captions. We further exploit the state of the art pre-trained image captioning and object recognition networks to annotate our images and show the limitations of existing works. Furthermore , in order to evaluate the quality of the generated captions, we propose a new image captioning metric, object based Semantic Fidelity (SF). Existing image cap-tioning metrics can evaluate a caption only in the presence of their corresponding annotations; however, SF allows evaluating captions generated for images without annotations, making it highly useful for real life generated captions.
Fichier principal
Vignette du fichier
ICLR2020.pdf (6.65 Mo) Télécharger le fichier
Origine : Fichiers produits par l'(les) auteur(s)
Loading...

Dates et versions

hal-02864865 , version 1 (11-06-2020)

Identifiants

  • HAL Id : hal-02864865 , version 1

Citer

Pranav Agarwal, Alejandro Betancourt, Vana Panagiotou, Natalia Díaz-Rodríguez. Egoshots, an ego-vision life-logging dataset and semantic fidelity metric to evaluate diversity in image captioning models. ICLR 2020 - 8th International Conference on Learning Representations, Apr 2020, Addis Ababa / Virtual, Ethiopia. ⟨hal-02864865⟩
89 Consultations
49 Téléchargements

Partager

Gmail Facebook X LinkedIn More