Deep Inside Visual-Semantic Embeddings

Abstract

In this thesis, we aim to further advance image representation and understanding. Revolving around Visual Semantic Embedding (VSE) approaches, we explore different directions: First, we present relevant background in Chapter 2, covering images and textual representation and existing multimodal approaches. Then in Chapter 3 we propose novel architectures further improving retrieval capability of VSE. In Chapter 4 we extend VSE models to novel applications and leverage embedding models to visually ground semantic concept. Finally, in Chapter 5 we delve into the learning process and in particular the loss function.

Links

Paper

Citation

If you found this work useful, please cite the associated paper:

@inproceedings{engilbergeThesis2020,
  title = {Deep Inside Visual-Semantic Embeddings},
  author = {Engilberge, Martin},
  year = {2020}
}