Show, Tell and Discriminate: Image Captioning by Self-retrieval with Partially Labeled Data
This addresses the issue of stereotyped captions in image captioning for AI applications, offering an incremental improvement by leveraging unlabeled data without extra annotations.
The paper tackled the problem of generating discriminative captions for images by proposing a self-retrieval module as training guidance, which improved captioning performance on COCO and Flickr30k datasets.
The aim of image captioning is to generate captions by machine to describe image contents. Despite many efforts, generating discriminative captions for images remains non-trivial. Most traditional approaches imitate the language structure patterns, thus tend to fall into a stereotype of replicating frequent phrases or sentences and neglect unique aspects of each image. In this work, we propose an image captioning framework with a self-retrieval module as training guidance, which encourages generating discriminative captions. It brings unique advantages: (1) the self-retrieval guidance can act as a metric and an evaluator of caption discriminativeness to assure the quality of generated captions. (2) The correspondence between generated captions and images are naturally incorporated in the generation process without human annotations, and hence our approach could utilize a large amount of unlabeled images to boost captioning performance with no additional laborious annotations. We demonstrate the effectiveness of the proposed retrieval-guided method on COCO and Flickr30k captioning datasets, and show its superior captioning performance with more discriminative captions.