Sentimentalized Visual Captioning using Attentional Autoencoders

Initially titled neural image sentiment comparison, the work here aims to extend previous work on visual sentiment analysis and captioning by proposing novel approaches to sentiment classification and caption combining using updates and noun-forcing. For further details, please read our paper here. Also note that no model checkpoints are included as part of the repository due to file size limits. If you would like access to the checkpoints, please email us a request at [email protected].