Is An Image Worth Five Sentences? A New Look into Semantics for Image-Text Matching

Ali Furkan Biten, Andres Mafla, Lluis Gomez, Dimosthenis Karatzas

Producció científica: Capítol de llibreCapítolRecercaAvaluat per experts

13 Cites (Scopus)

Resum

The task of image-text matching aims to map representations from different modalities into a common joint visual-textual embedding. However, the most widely used datasets for this task, MSCOCO and Flickr30K, are actually image captioning datasets that offer a very limited set of relation-ships between images and sentences in their ground-truth annotations. This limited ground truth information forces us to use evaluation metrics based on binary relevance: given a sentence query we consider only one image as relevant. However, many other relevant images or captions may be present in the dataset. In this work, we propose two metrics that evaluate the degree of semantic relevance of retrieved items, independently of their annotated binary relevance. Additionally, we incorporate a novel strategy that uses an image captioning metric, CIDEr, to define a Semantic Adaptive Margin (SAM) to be optimized in a standard triplet loss. By incorporating our formulation to existing models, a large improvement is obtained in scenarios where available training data is limited. We also demonstrate that the performance on the annotated image-caption pairs is maintained while improving on other non-annotated relevant items when employing the full training set. The code for our new metric can be found at github.com/furkanbiten/ncs_metric and the model implementation at github.com/andrespmd/semantic_adaptive_margin.

Idioma originalAnglès
Títol de la publicacióProceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022
EditorInstitute of Electrical and Electronics Engineers Inc.
Pàgines2483-2492
Nombre de pàgines10
ISBN (electrònic)9781665409155
DOIs
Estat de la publicacióPublicada - 2022

Sèrie de publicacions

NomProceedings - 2022 IEEE/CVF Winter Conference on Applications of Computer Vision, WACV 2022

Fingerprint

Navegar pels temes de recerca de 'Is An Image Worth Five Sentences? A New Look into Semantics for Image-Text Matching'. Junts formen un fingerprint únic.

Com citar-ho