English

Full-Network Embedding in a Multimodal Embedding Pipeline

Computer Vision and Pattern Recognition 2017-08-10 v2 Computation and Language Neural and Evolutionary Computing

Abstract

The current state-of-the-art for image annotation and image retrieval tasks is obtained through deep neural networks, which combine an image representation and a text representation into a shared embedding space. In this paper we evaluate the impact of using the Full-Network embedding in this setting, replacing the original image representation in a competitive multimodal embedding generation scheme. Unlike the one-layer image embeddings typically used by most approaches, the Full-Network embedding provides a multi-scale representation of images, which results in richer characterizations. To measure the influence of the Full-Network embedding, we evaluate its performance on three different datasets, and compare the results with the original multimodal embedding generation scheme when using a one-layer image embedding, and with the rest of the state-of-the-art. Results for image annotation and image retrieval tasks indicate that the Full-Network embedding is consistently superior to the one-layer embedding. These results motivate the integration of the Full-Network embedding on any multimodal embedding generation scheme, something feasible thanks to the flexibility of the approach.

Keywords

Cite

@article{arxiv.1707.09872,
  title  = {Full-Network Embedding in a Multimodal Embedding Pipeline},
  author = {Armand Vilalta and Dario Garcia-Gasulla and Ferran Parés and Eduard Ayguadé and Jesus Labarta and Ulises Cortés and Toyotaro Suzumura},
  journal= {arXiv preprint arXiv:1707.09872},
  year   = {2017}
}

Comments

In 2nd Workshop on Semantic Deep Learning (SemDeep-2) at the 12th International Conference on Computational Semantics (IWCS) 2017

R2 v1 2026-06-22T21:02:23.111Z