Areas of Attention for Image Captioning - Archive ouverte HAL Access content directly
Conference Papers Year : 2017

## Areas of Attention for Image Captioning

Marco Pedersoli
• Function : Author
Thomas Lucas
• Function : Author
Cordelia Schmid
• Function : Author
Jakob Verbeek

#### Abstract

We propose Areas of Attention'', a novel attention-based model for automatic image captioning. Our approach models the dependencies between image regions, caption words, and the state of an RNN language model, using three pairwise interactions. In contrast to previous attention-based approaches that associate image regions only to the RNN state, our method allows a direct association between caption words and image regions. During training these associations are inferred from image-level captions, akin to weakly-supervised object detector training. These associations help to improve captioning by localizing the corresponding regions during testing. We also propose and compare different ways of generating attention areas: CNN activation grids, object proposals, and spatial transformers nets applied in a convolutional fashion. Spatial transformers give the best results. They allow for image specific attention areas, and can be trained jointly with the rest of the network. Our attention mechanism and spatial transformer attention areas together yield state-of-the-art results on the MSCOCO dataset.

### Dates and versions

hal-01428963 , version 1 (06-01-2017)
hal-01428963 , version 2 (25-08-2017)

### Identifiers

• HAL Id : hal-01428963 , version 2
• DOI :

### Cite

Marco Pedersoli, Thomas Lucas, Cordelia Schmid, Jakob Verbeek. Areas of Attention for Image Captioning. ICCV - International Conference on Computer Vision, Oct 2017, Venice, Italy. pp.1251-1259, ⟨10.1109/ICCV.2017.140⟩. ⟨hal-01428963v2⟩

### Export

BibTeX TEI Dublin Core DC Terms EndNote Datacite

1073 View